JP2000115225A - Network changeover method applied to network system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To switch a network without the use of a special communication application. SOLUTION: In the system provided with computers 10A, 10B that connect with LANs 11, 12 and that have respectively an NIC 101 to which addresses IP1, IP1' of a 1st system are set and an NIC 102 to which addresses IP2, IP2' of a 2nd system are set, the computers 10A, 10B set addresses IP, IP' shared by the LANs 11, 12 additionally to the NIC 101 to conduct communication between the computers. On the occurrence of a fault in the LAN 11 in this state or at the arrival of an instruction for LAN switching by the user, the computer 10A, 10B delete the IP, IP' from the NIC 101 and set the IP, IP' additionally to the NIC 102 to changeover the networks.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention comprises a plurality of computers connected to a multiplexed network (transmission path), and communication between the computers is performed using one of the multiplexed networks. The present invention relates to a network switching method in a network system.

[0002]

2. Description of the Related Art In a system in which communication between computers is performed via a network (transmission path), when a failure occurs on the network, communication between the computers is hindered. Therefore, in recent years, a network system (a LAN-resistant computer system) in which a network (LAN) is multiplexed so that communication between computers can be performed even when a failure occurs on the network has become mainstream.

[0003] A typical multiplexed network (multiplexed LAN, multi-system LAN) is a duplex LAN composed of a main LAN and a sub-LAN. The computer connected to this dual LAN has two connection ports (network interface) so that it can be connected to both LANs.
have. If a failure occurs on one LAN while communicating using the LAN of one system, the computer switches the LAN for communication to the other system by switching the connection port. Here, the connection port is switched by the computer switching an address used for communication by a communication application to an alternative address.

[0004] However, in the above-described method, the destination part of the communication packet (the destination address) needs to be changed by the communication application, so that the LAN can be switched only by a specially created communication application.

[0005] As shown in FIG. 44, computers 100A and 100B are connected to a primary LAN1.
In the fault-tolerant computer system connected to the secondary network 11 and the secondary LAN 112, the computer 100A uses TCP / IP (transmis- sion).
Sion control protocol / int
(Ethernet protocol), and the computer 100B executes a TCP / IP client program. The IP address of the connection port of the computer 100A with the main LAN 111 is X1, the IP address of the connection port with the slave LAN 112 is Y1, and the IP address of the connection port of the computer 100B with the main LAN 111 is X1.
2. It is assumed that the IP address of the connection port with the slave LAN 112 is Y2.

Here, computers 100A and 10A
The default contents of the routing table of 0B are as follows: * Computer 100A Destination = X2, Gateway = X2 Destination = Y2, Gateway = Y2 * Computer 100B Destination = X1, Gateway = X1 Destination = Y1, Gateway = Y1 I have. In addition, Destination =
A set of i, Gateway = j is connected to the address j of the communication destination computer (NIC in which is set) (inquiry of the route) in order to transmit (connect to i) to the address i. The route information indicates that it is necessary to communicate. j (Gateway) is also called a destination (or an inquiry destination).

The main LAN in such a system
Computers 100A, 10 when 111 can communicate
A description will be given of a conventional technique relating to a connection between 0B and a connection between the computers 100A and 100B when the main LAN 111 cannot communicate.

(1) Connection When Main LAN 111 Can Communicate While the main LAN 111 is communicable, the computer (client) 100B sends the computer (server)
When connecting to (X1) of the main LAN 100A, even if the socket (interface for transmitting and receiving data by inter-process communication on the transport layer in the TCP / IP network) is not explicitly bound, the main LAN
It is automatically bound (as a source address) to the address X2 of the connection port with the 111 side. That is, X2 is the source address of TCP / IP.

In this state, if the main LAN 111 becomes unable to communicate, Destination = X2, Gateway = Y2 is added to the routing table of the computer 100A, and Destination = X1, Gateway = Y1 is added to the routing table of the computer 100B. Is added. These show a detour route (a detour route) to the destination address when the main LAN 111 becomes unable to communicate.

In this case, the computer (client)
100B intends to transmit to X1, but because the routing table has Destination = X1 and Gateway = Y1, the request can be transmitted using the slave LAN 112. Therefore, communication can be continued while maintaining a TCP / IP session.

On the other hand, the computer (server) 100A
Attempts to return a response to the source address X2 of TCP / IP, but since the routing table has Destination = X2 and Gateway = Y2, the response can be returned using the slave LAN 112. Therefore, communication can be continued while maintaining a TCP / IP session.

(2) When the connection is made when the main LAN 111 cannot communicate In the first state in which the main LAN 111 cannot communicate and the subordinate LAN 112 can communicate, the computer (client) 100B to the computer (server) 100A
When connecting to (X1), even if the socket is not explicitly bound, the communicable slave LAN1
It is automatically bound (as a source address) to the address Y2 of the connection port with the 12 side. That is, Y2 is T
The source address of the CP / IP.

From the first state, the slave LAN 1
12 cannot communicate and the main LAN 111 can communicate.
In this state, since the main LAN 111 can communicate, the routing tables of the computers 100A and 100B have default contents.

In general, the default routing table uses a LAN with a transmission address. For this reason, when the state changes from the first state to the second state, the computer (client) 100B
However, transmission can be performed using the main LAN 111 according to Destination = X1 and GatewayX1 in the default routing table. Therefore, communication can be continued while maintaining a TCP / IP session.

On the other hand, the computer (server) 100A
Is bound to Y2 on the slave LAN 112 side, and thus attempts to return a response to the TCP / IP source address Y2. However, in the default routing table, since the destination LAN 112 (communication disabled) is used because Destination = Y2 and Gateway = Y2, no response can be made. Therefore, TCP / I
Communication of P cannot be continued.

As described above, conventionally, connections between computers are made in a state where the main LAN (main transmission line) is not communicable and the subordinate LAN (subordinate transmission line) is communicable.
If the AN is communicable and the slave LAN changes to a non-communicable state, communication cannot be continued.

[0017]

As described above, in a conventional network system (a LAN-resistant computer system) in which a plurality of computers are connected to a multiplexed LAN (multiplexed network), each computer is used for communication. In order to switch the network (LAN, transmission path), it is necessary to change the destination part of the communication packet by the communication application, so that there is a problem that the LAN can be switched only by a specially created communication application.

Further, the conventional network system (LA resistant)
N failure computer system), the computers are connected in a state where the main LAN is not communicable and the slave LAN is communicable, and then the main LAN is communicable and the slave LA is connected.
There is also a problem that communication cannot be continued when N changes to a communication disabled state.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a LAN switching method in a network system in which a network can be switched without using a special communication application.

Another object of the present invention is to provide a LAN switching method in a network system in which communication can always be continued as long as a communicable network exists even if one of the multiplexed networks becomes unable to communicate. It is in.

[0021]

SUMMARY OF THE INVENTION The present invention relates to a computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network, A first system network interface to which a second system address is set, and a second system network interface which forms a connection port to a second system network and has a second system address set thereto. It is possible to provide a plurality of computers and perform communication between the respective computers using the corresponding first or second network using an address set in the network interface of the first or second network. In a possible network system (a computer system that is resistant to LAN failure), a first network interface is used for each computer. By additionally setting a common shared address unique to the own computer and common to the first system and the second system in the network, communication between the computers is performed using the shared address. When a failure occurs in the network of one system (the network of one system) or when there is an instruction to switch the network by the user, each computer switches from the network interface of the first system (the network interface of one system). The network is switched by deleting the shared address and executing a process of adding and setting the shared address to the second system network interface (the other system network interface).

In such a configuration, although the computers (LANs) to be used are switched, the computers can communicate with each other using the same shared address as before the occurrence of the failure. That is, since communication can be continued without changing the destination part (the destination address) of the communication packet, a special communication application is not required.
Further, in the state after the network switching, communication using the second system address is possible in addition to communication using the shared address. Also, before the network switch,
Communication is possible with all of the addresses of the first system, the addresses of the second system, and the shared addresses, and both networks of the two systems can be used.

If the network interfaces of the first system and the second system are logically incorporated in the system, the network of the first system (the network of one system) as described above. In the case where a failure occurs in the network or when a user issues an instruction to switch the network, instead of deleting the shared address from the network interface of the first system (the network interface of the one system) in each computer, Even when the network interface of the first system is separated from the system, the same effect as that of deleting the shared address can be obtained.

Further, according to the present invention, when a failure occurs in the first system network (one system network) or when there is an instruction to switch the network by the user, each computer can execute the first system network. Interface (network interface of one system)
, The address of the first system (the address of the one system) is deleted, and the address of the first system (the address of the one system) is replaced with the network interface of the second system (the network interface of the other system). It is also characterized in that a network can be switched without using a shared address by executing a process of additionally setting the network.

In such a configuration, each computer can communicate with each other using the same system address as before the failure, although the network used is switched. That is, since communication can be continued without changing the destination part (the destination address) of the communication packet, a special communication application is not required.

If the network interfaces of the first system and the second system are logically incorporated in the system, the network of the first system (the network of one system) as described above. When a failure occurs in the network or when a user issues a network switching instruction, each computer sends the first system address (the one system network interface) from the first system network interface (one system network interface). Instead of deleting the address of the first system, the network interface of the first system (the network interface of the one system) is disconnected from the system, and the address of the first system (the address of the one system) is deleted. An equivalent effect can be obtained.

In addition, when a failure occurs in the first network (one of the networks) or when a user issues a network switching instruction,
The one computer that has detected the failure or given the instruction to switch the network performs the above-described operation on the network interface of the first system and the second system with respect to all the other computers by using the second computer. It is also possible to adopt a configuration in which an instruction is given using the network of the second system (the network of the other system), while the operation on the network interface of the first system and the second system described above is performed within the own computer. . However, one computer needs a function of issuing the above instruction (an application program for realizing the instruction), and all other computers need a function of receiving and executing the above instruction.

Similarly, when a failure occurs in the first network (one of the networks) or when a user issues a network switching instruction, the failure is detected or a network switching instruction is given. One computer performs the above-mentioned operations on the network interfaces of the first system and the second system with respect to all the other computers using the network of the second system (the network of the other system). Alternatively, it is also possible to adopt a configuration in which the above-mentioned first system and second system network interfaces are operated in the own computer while being executed by remote processing. However,
One computer needs a function (an application program for realizing the function of executing a remote process by logging in to all the other computers).

According to the present invention, there is also provided a computer connected to a first system network and a second system network, the first system network interface forming a connection port with the first system network. Which one of the first and second network interfaces of the own computer is used when communicating with the designated address, and the second system network interface forming a connection port with the second system network Network system that includes a plurality of computers each having a routing table in which an information entry indicating the information is registered, and is capable of performing communication between the computers using the first system or the second system network (LAN-resistant system) Computer system), a first network (one network) ), When a failure occurs, or when a user issues an instruction to switch networks, each of the above-mentioned computers is used to determine the address of the first network (the one network). The information entry indicating that the second system network interface (the other system network) is to be used for the communication with the address of the first system network (the one system network) in the routing table. Network switching is performed by adding the information to a position searched before other information entries related to the network.

In such a configuration, each computer can communicate with each other using the same system address as before the failure, although the network to be used is switched. That is, since communication can be continued without changing the destination part (the destination address) of the communication packet, a special communication application is not required.

It should be noted that, after the network is switched by the addition of the information entry, if the failure is resolved or if the user issues an instruction to restore the network switching, the added information entry is deleted from the routing table. Good to do. Network switching by adding or deleting information entries is the first
It is also possible to realize mutual switching by simultaneously executing the network of the second system and the network of the second system.

Further, when a failure occurs in the first network (one of the networks) or when there is a network switching instruction from the user, the failure is detected or the network switching instruction is given. One computer instructs all the other computers to execute the above-described operation on the routing table using the second system network (the other system network). It is also possible to adopt a configuration for performing an operation on the routing table described above.

Similarly, when a failure occurs in the first system network (one of the networks) or when a user issues a network switching instruction, the failure is detected or the network switching instruction is given. While the one computer performs the operation on the routing table for all the other computers by remote processing using the network of the second system (the network of the other system), Within the above, it is also possible to adopt a configuration in which the above-mentioned operation on the routing table is performed.

The present invention also relates to a computer connected to a multiplexed N-system network, wherein the computer has a connection port to each of the N-system networks, and has N addresses each of which has a corresponding system address. Computer having a network interface and a routing table in which an information entry indicating a route on the communication destination computer side to the designated address is registered, and each computer is provided using any of the N-type networks. In a network system capable of performing communication between computers (LAN-resistant computer system), the above-mentioned N-system networks are treated equally, and if there is a network that cannot communicate, the network that cannot communicate is indicated. The latest representing a detour to an address with a network address part By adding one information entry to the routing table at a position searched before the information entry representing the standard route to the address in the routing table, and deleting any information entries representing other detour routes, The network is switched.

In such a configuration, when there is a network that cannot communicate, there is no information entry indicating the previous detour route (detour route) to the address in the routing table, and the address is not stored in the routing table. , That is, an information entry representing the latest detour path that matches the current state of the network always exists at a position searched before an information entry representing a standard path to that address. Therefore, when performing communication to an address having a network address assigned to an uncommunicable network, searching the routing table to obtain the route information necessary for communication to that address results in the latest Can be obtained without fail, and communication can be performed correctly in accordance with the information entry. For this reason, for example, in a system provided with a duplicated network, connections between computers are made in a state where one system network cannot communicate and the other system network can communicate, and then one system network communicates. Communication can be continued even if the network of the other system is changed to a state where communication is not possible.

[0036]

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

[First Embodiment] FIG. 1 is a block diagram showing the configuration of a LAN-resistant computer system (network system) according to a first embodiment of the present invention.

In the figure, a first LAN (transmission line) 11 and a second LAN 12 constituting a duplicated LAN (transmission path) include a plurality of computers, for example, computers 10A and 10B. Is connected. The computers 10A and 10B communicate using the LAN 11 or 12.

Each of the computers 10A and 10B includes a first system network interface (hereinafter, referred to as NIC) 101 serving as a connection port to the first system LAN 11.
A second system NIC (network interface) 102 serving as a connection port to the second system LAN 12,
An input / output processor 103 connected to the NICs 101 and 102 is provided.

The input / output processor 103 has an address storage unit 104 for storing an address common to the first system and the second system (hereinafter, referred to as a shared address), and a LAN switching unit 105.

Address storage unit 10 of computer 10A
4, the shared address IPA is stored in the address storage unit 104 of the computer 10B, and the shared address IPB is stored in the address storage unit 104 of the computer 10B. Shared address IPA, I
PB is an IP address and is composed of a network address and a host address. Here, the network address is common to both IPA and IPB, and LA
N11 and N12 are assigned a common network address for the IPA and IPB. On the other hand, the host address differs between IPA and IPB. That is, unique host addresses are assigned to the computers 10A and 10B. Thus, the shared addresses IPA, IPB
Is an address unique to the computers 10A and 10B.

The LAN switching unit 105 has a function of setting an address shared by the first system (the LAN 11) and the second system (the LAN 12) in the NICs 101 and 102, and a system of the NICs 101 and 102 which are logically incorporated. And logically separated NIC 101
And a NIC disconnection / incorporation function for incorporation into the system of FIG.

Next, the LAN switching of each computer in the system having the configuration shown in FIG. 1 will be described with reference to the flowchart of FIG. 2 and the system state diagram of FIG.

The LAN 11 of the first system and the LA of the second system
In all the computers (here, computers 10A and 10B) connected to N12, the shared addresses IPA and IA stored in the own address storage unit 104 are used by the shared address setting function of the LAN switching unit 105.
PB is set in advance in its own first system NIC 101 and second system NIC 102 (step S1).

Next, in the computers 10A and 10B, L
The NIC switching function of the AN switching unit 105 logically incorporates its own first system NIC 101 into the system (step S2). FIG. 3A shows the state of the system at this time. Here, it is shown that the NIC connected to the LAN by a solid line is logically incorporated in the system, and that the NIC connected to the LAN by a broken line is logically separated from the system. This figure 3
In the state of (a), the computers 10A and 10B
Using the LAN 11 of the first system, the shared address IPA,
Communicate by IPB.

As described above, while the computers 10A and 10B are communicating with each other using the LAN 11 of the first system, a failure has occurred in the LAN 11 or an instruction to switch the LAN has been given by a user operation. Shall be. In this case, the computers 10A and 10B
With the NIC disconnection function of the LAN switching unit 105,
The first NIC 101 of the own system is logically separated from the system (step S3), and then the second NIC 102 of the second system is logically incorporated into the system by the NIC incorporating function of the LAN switching unit 105 (step S4). . The state of the system at this time is shown in FIG.

As described above, the switching of the LAN can be easily realized. In this state, the computer 10A,
10B indicates that the LAN to be used is from LAN11 to LAN12.
, But the same shared address I as before the failure occurred
Communication can be performed mutually using PA and IPB. That is, since communication can be continued without changing the destination part (the destination address) of the communication packet, a special communication application is not required.

[Second Embodiment] Next, a second embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

A feature of this embodiment is that the first embodiment sets a shared address in advance to both the NICs 101 and 102, whereas the first embodiment sets a shared address only to the NIC 101 of the first system, for example. is there. Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart of FIG. 4 and the system state diagram of FIG.

The LAN 11 of the first system and the LA of the second system
In all the computers (computers 10A and 10B in this case) connected to N12, the shared address setting function of the LAN switching unit 105 causes the shared address I
PA and IPB are set in advance in the own first system NIC 101 (step S11).

Next, in the computers 10A and 10B, L
The NIC 101 is logically incorporated into the system by the NIC incorporating function of the AN switching unit 105 (step S12). FIG. 5A shows the state of the system at this time. In the state of FIG. 5A, the computer 10A
And 10B communicate using the first system LAN 11 with the shared addresses IPA and IPB.

As described above, while the computers 10A and 10B are communicating with each other using the LAN 11 of the first system, a failure has occurred in the LAN 11 or an instruction to switch the LAN has been given by a user operation. Shall be. In this case, the computers 10A and 10B
With the NIC disconnection function of the LAN switching unit 105,
The NIC 101 of its own first system is logically disconnected from the system (step S13). Then the computer 10
A and 10B use the shared address setting function of the LAN switching unit 105 to set the shared addresses IPA and IPB in their own second system NIC 102 (step S1).
4) After that, the NIC is incorporated by the NIC built-in function.
C102 is logically incorporated into the system (step S1)
5). The state of the system at this time is shown in FIG.

As described above, the switching of the LAN can be easily realized.

[Third Embodiment] Next, a third embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

The feature of this embodiment is that the LAN switching unit 1
05 has a function of deleting a shared address set in the NIC instead of a function of logically separating the NIC from the system. Hereinafter, LAN switching when the LAN switching unit 105 is provided with a shared address deletion function, more specifically, LAN when the shared address deletion processing is applied instead of the NIC separation in the first embodiment. The switching will be described with reference to the flowchart of FIG. 6 and the system state diagram of FIG.

First, in all computers (here, computers 10A and 10B) connected to the first system LAN 11 and the second system LAN 12, the shared address setting function of the LAN switching unit 105 causes the shared address IPA. , IPB and its own first system NIC 101
After setting in advance to the NIC 102 of the second system (step S21), the NIC 10
1 is logically incorporated into the system (step S22).
FIG. 7A shows the state of the system at this time. Up to this point, the operation is the same as in the first embodiment.

Next, in the state of FIG. 7A, it is assumed that a failure of the LAN 11 has occurred or a LAN switching instruction has been given by a user operation. In this case, the computers 10A and 10B use the shared address deletion function of the LAN switching unit 105 to execute their own first system NIC 10
The addresses IPA and IPB set to 1 are deleted (step S23), and then the LAN switching unit 10 is deleted.
5 NIC disconnection function of its own second system
C102 is logically incorporated into the system (step S2).
4). FIG. 7B shows the state of the system at this time.

As described above, the switching of the LAN can be easily realized.

[Fourth Embodiment] Next, a fourth embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

The feature of this embodiment is that the shared address deletion processing used in the third embodiment is applied instead of the NIC separation in the second embodiment.
Hereinafter, LAN switching according to the present embodiment will be described with reference to the flowchart in FIG. 8 and the system state diagram in FIG.

The LAN 11 of the first system and the LA of the second system
In all the computers (computers 10A and 10B in this case) connected to N12, the shared address setting function of the LAN switching unit 105 causes the shared address I
PA and IPB are set in its own first system NIC 101 (step S31), and then the LAN switching unit 1
The NIC 101 is logically incorporated into the system by the NIC incorporating function 05 (step S32). FIG. 9A shows the state of the system at this time. Up to this point, the operation is the same as in the second embodiment.

Next, in the state of FIG. 9A, it is assumed that a failure of the LAN 11 has occurred or a LAN switching instruction has been given by a user operation. In this case, the computers 10A and 10B use the shared address deletion function of the LAN switching unit 105 to execute their own first system NIC 10
The addresses IPA and IPB set to 1 are deleted (step S33). Then, the computers 10A, 10
In B, the shared address setting function of the LAN switching unit 105 sets the shared addresses IPA and IPB to its own second address.
(Step S34), and then the NIC 102 is logically incorporated into the system by the NIC incorporating function (step S35). FIG. 9B shows the state of the system at this time.

As described above, LAN switching can be easily realized.

[Fifth Embodiment] Next, a fifth embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

This embodiment is characterized in that the NICs 101 and 1
02 and a function to set a shared address in one of
By using the LAN switching unit 105 having a function of incorporating both 01 and 102 into the system and a function of deleting the shared address, a configuration is adopted in which the shared address is set only to the NIC 101 or 102 of the system actually used for communication. On the point. Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart of FIG. 10 and the system state diagram of FIG.

The LAN 11 of the first system and the LA of the second system
In all the computers (computers 10A and 10B in this case) connected to N12, the shared address setting function of the LAN switching unit 105 causes the shared address I
PA and IPB are set in its own first system NIC 101 (step S41), and then the LAN switching unit 1
With the NIC incorporating function 05, both the NICs 101 and 102 are logically incorporated into the system (step S42). FIG. 11A shows the state of the system at this time.
Shown in

Next, in the state of FIG.
It is assumed that a failure has occurred or a LAN switching instruction has been issued by a user operation. In this case, each of the computers 10A and 10B uses the shared address deletion function of the LAN switching unit 105 to execute its own NIC1 of the first system.
01 are deleted (step S43), and then the LAN switching unit 10
5, the shared address setting function
, IPB are set to the own NIC 102 of the second system (step S44). The state of the system at this time is shown in FIG.

As described above, the switching of the LAN can be easily realized.

[Sixth Embodiment] Next, a sixth embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

In the present embodiment, as in the prior art, an IP address (including a unique network address) unique to the system (the LAN 11, 12) corresponding to the NIC 101, 102 of each computer 10i (i = A, B) is specified in advance. The NICs 101 and 102 are logically incorporated in the system.

This embodiment is characterized in that not only the LANs 11 and 12 are assigned unique network addresses but also a common network address is assigned to both LANs 11 and 12.
A, 10B NIC 101 or 102 has a common network address, that is, LA
The point is that IP addresses (hereinafter, referred to as shared addresses) IP and IP 'indicating N11 and N12 are additionally set.
The host address of the address IP indicates the computer 10A, and the host address of the address IP 'indicates the computer 10B. Address storage unit 1 of computer 10A
04 stores the address IP and the computer 10B
The address storage unit 104 stores an address IP '. In this respect, the address storage unit 104 of the computers 10A and 10B applied in the first to fifth embodiments.
And different.

Here, the NIC 1 of the computer 10A
01 and 102 have IP addresses IP1 and IP2, respectively.
2 is set in advance and the NIC 10 of the computer 10B is set.
IP addresses IP1 'and IP1
2 'is set in advance. Address IP1, IP1 '
Network addresses indicate the LAN 11, and the host addresses indicate the computers 10A and 10B. On the other hand, the network addresses of the addresses IP12 and IP2 'both indicate the LAN 12, and the host addresses indicate the computers 10A and 10B.

Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart of FIG. 12 and the system state diagram of FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 10A and 10B, the address setting function (shared address setting function) of the LAN switching unit 105.
Thereby, the shared addresses IP and IP 'are additionally set to the own first system NIC 101 (step S51). In this state, the computers 10A and 10B have their own N
From the IC 101 to the shared address I using the LAN 11
Communication can be performed by P, IP '. The computers 10A and 10B can communicate with each other using the LAN 11 by the addresses IP1 and IP1 '.
2, it is also possible to communicate by the addresses IP2 and IP2 '. FIG. 13 shows the state of the system at this time.
(A).

Next, in the state of FIG.
It is assumed that a failure has occurred or a LAN switching instruction has been issued by a user operation. In this case, the computers 10A and 10B delete the shared addresses IP and IP 'set in their own first system NIC 101 by the address deletion function (shared address deletion function) of the LAN switching unit 105 (step S52). Then, the address setting function (shared address setting function) of the LAN switching unit 105 causes the shared addresses IP and IP 'to be set.
Is additionally set to the own NIC 102 of the second system (step S53). That is, the computers 10A and 10B
Moves the shared addresses IP and IP 'from the NIC 101 to the NIC 102 in the own computer. The state of the system at this time is shown in FIG.

As described above, LAN switching can be easily realized. In this state, the computer 10A,
10B indicates that the LAN to be used is from LAN11 to LAN12.
, But the same shared address I as before the failure occurred
They can communicate with each other using P and IP '. That is, since communication can be continued without changing the destination part (the destination address) of the communication packet, a special communication application is not required. Also, the addresses IP2 and IP2
Communication can also be performed using 2 '.

In the present embodiment, when the LAN 11 is communicable, all the addresses (including the shared address) set in the NICs 101 and 102 can communicate, and the first system LAN 11 and the second LA of system 2
Communication is possible with both systems of N12.

[Seventh Embodiment] In the sixth embodiment, when a failure such as a LAN occurs, or when a user issues an instruction to switch LANs, all the computers in the system execute the above steps. S52,
S53 has been described as being executed autonomously. On the other hand, in the present embodiment, one computer that has detected a failure such as a LAN or a computer that has received a LAN switching instruction from a user executes the above-described steps S52 and S53 on all other computers (LAN
The execution of the switching process is instructed using a normal LAN, while the LAN switching process is executed in the own computer.

FIG. 14 is a block diagram showing a configuration of a LAN-resistant computer system according to the present embodiment (seventh embodiment of the present invention), and the same reference numerals are assigned to the same parts as in FIG.

In the figure, a first LAN (transmission line) 11 and a second LAN 12 constituting a duplicated LAN (transmission path) include (computer 10 in FIG. 1).
Computers 20A and 20 (corresponding to A and 10B)
B is connected.

The computer 20A has the NICs 101, 1
02 and an input / output processor 103 (having an address storage unit 104 and a LAN switching unit 105), and a switching instruction unit 200. The switch instructing unit 200 normally executes the LAN switch processing on another computer (the computer 10B in this case) by detecting a failure in the LAN or the like or receiving an instruction to switch the LAN from the user. Na L
Instructions are given using an AN. The switching instruction unit 200 is realized by an application program and a computer 20A (not shown, which is a central part of the computer) that executes the application program.

On the other hand, similarly to the computers 10A and 10B in FIG.
02 and the input / output processor 103, or a configuration in which a switching instruction unit 200 is further added as in the case of the computer 20A. Here, it is assumed that the computer 20B has the same configuration as the computers 10A and 10B.

In the configuration shown in FIG. 14, the computer 2
OA is a server (TCP / IP server), and the computer 20B is a client (TCP / IP client).

Next, the switching of the LAN of each computer in the system having the configuration shown in FIG. 14 will be described with reference to the flowchart of FIG. 15 and the system state diagram of FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 20A and 20B, the shared address setting function of the LAN switching unit 105 causes the shared addresses IP and IP. 'Is additionally set to its own first system NIC 101 (steps S61 and S71). In this state, the computers 20A and 20B can communicate with the shared addresses IP and IP 'from the own NIC 101 using the LAN 11. FIG. 16A shows the state of the system at this time.

Next, a failure of the LAN 11 occurs in the state of FIG. 16A, and the computer 20A detects that a failure has occurred, or the
As a result of inputting the AN switching instruction, it is assumed that the switching instruction is received by the computer 20A.

In this case, in the computer 20A, the normal LAN 12 (and IP2, IP2 ') is used by the switching instruction unit 200 with respect to another computer (here, the computer 20B) as shown in FIG. And the LAN (by the same method as in the sixth embodiment)
(Step S62).
Also, the computer 20A deletes the shared address IP set in its own first system NIC 101 by the shared address deletion function of the LAN switching unit 105, similarly to the sixth embodiment (step S63). Thereafter, the shared address setting function of the LAN switching unit 105 transfers the shared address IP to the NI of its own second system.
Additional settings are made for C102 (step S64).

On the other hand, in the computer 20B in which the execution of the LAN switching process is instructed by the switching instructing unit 200 in the computer 20A, the first addressing NIC of the computer 20B is operated by the shared address deleting function of the LAN switching unit 105.
Delete the shared address IP 'set in 101 (step S72), and after that, the LAN switching unit 10
5, the shared address IP '
Is additionally set to its own second system NIC 102 (step S73). FIG. 16 shows the state of the system at this time.
It is shown in (c).

As described above, the switching of the LAN can be easily realized. In this state, the computer 20A,
20B indicates that the LAN to be used is from LAN11 to LAN12.
, But the same shared address I as before the failure occurred
They can communicate with each other using P and IP '.

[Eighth Embodiment] In the seventh embodiment, one computer that has detected a failure such as a LAN or a computer that has received a LAN switching instruction from a user is connected to all other computers. LAN
The description has been given assuming that the execution of the switching process is instructed using a normal LAN. On the other hand, in the present embodiment, L
One computer that has detected a failure such as an AN, or the computer itself that has received the LAN switching instruction from the user, uses the normal LAN for all the other computers and uses the above-described steps S52 and S53 (the shared address). (Deletion and addition of a shared address) are executed by remote processing.

FIG. 17 is a block diagram showing a configuration of a LAN-resistant computer system according to the present embodiment (eighth embodiment of the present invention). The same parts as those in FIG. 1 are denoted by the same reference numerals.

In the figure, the first system LAN (transmission path) 11 and the second system LAN 12 constituting the duplicated LAN (transmission path) include (the computer 10 in FIG. 1).
Computers 30A and 30 (corresponding to A and 10B)
B is connected.

The computer 30A includes the NICs 101 and 1
02, and an input / output processor 103 (having an address storage unit 104 and a LAN switching unit 105), in addition to another computer address storage unit 301 and another computer LAN switching unit 302.

The other computer address storage section 301 in the computer 30A stores a shared address used by another computer (here, a shared address IP 'used by the computer 30B). Other computer LAN switching unit 302 in computer 30A
Executes the LAN switching in steps S52 and S53 (deletion of the shared address and additional setting of the shared address) by remote processing using the normal LAN for all the other computers. The other-computer LAN switching unit 302 is realized by an application program and a computer 30A (not shown, which forms the center of the computer) that executes the program.

On the other hand, the computer 30B has a configuration including the NICs 101 and 102 and the input / output processor 103 similarly to the computer 10A in FIG. 103 and the switch instructing unit 200, the NIC 10 is similar to the computer 30A.
1, 102, the input / output processor 103, the other computer address storage unit 301, and the other computer LAN switching unit 302.

In the configuration shown in FIG. 17, the computer 3
0A is a server (TCP / IP server), and the computer 30B is a client (TCP / IP client).

Next, the LAN switching of each computer in the system having the configuration of FIG. 17 will be described with reference to the flowchart of FIG. 18 and the system state diagram of FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 30A and 30B, the shared address setting function of the LAN switching unit 105 causes the shared addresses IP and IP. 'Is additionally set to its own first system NIC 101 (steps S81 and S91). In this state, the computers 30A and 30B can communicate with each other from the own NIC 101 using the LAN 11 with the shared addresses IP and IP '. FIG. 19A shows the state of the system at this time.

Next, a failure of the LAN 11 occurs in the state of FIG. 19A, and the computer 30A detects that a failure has occurred, or the
As a result of inputting the instruction to switch the AN, it is assumed that the switching instruction is received by the computer 30A.

In this case, in the computer 30A, the normal computer LA switching unit 302
Using N12 (and IP2, IP2 ′), a shared address that matches the shared address applied to the other computer (computer 30B) stored in the other computer address storage unit 301, that is, IP ′, is assigned to the other computer. NIC1 of the first system in (computer 30B)
01 by remote processing (step S8)
2). In the computer 30A, this step S
In parallel with the process 82, the LAN switching unit 105 in the own computer deletes the shared address IP from its own first system NIC 101 (step S83). FIG. 19B shows the state of the system at this time.

Subsequently, in the computer 30A, the normal LAN 1
2 (and IP2, IP2 '), the shared address IP' is changed to the computer 30B which is another computer.
Are additionally set to the NIC 102 of the second system by remote processing (step S84). The computer 30
In A, in parallel with this step S84, the LAN switching unit 105 in the own computer additionally sets the shared address IP to its own second NIC 102 (step S85). Figure 1 shows the state of the system at this time.
This is shown in FIG. 9 (c).

As described above, LAN switching can be easily realized. In this state, the computer 30A,
30B is used from LAN11 to LAN12.
, But the same shared address I as before the failure occurred
They can communicate with each other using P and IP '.

[Ninth Embodiment] Next, a ninth embodiment of the present invention will be described. FIG. 1 is referred to for the system configuration for convenience.

This embodiment is characterized in that the LAN switching unit 1
05, the NIC 10 (applied in the sixth embodiment)
In place of the function of deleting the shared address set in one of the NICs 1 and 102, a function of logically separating the NIC from the system is provided. Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart in FIG. 20 and the system state diagram in FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 10A and 10B, the shared address setting function of the LAN switching unit 105 causes the shared addresses IP and IP. 'Is additionally set to its own first system NIC 101 (step S101). FIG. 21A shows the state of the system at this time.

Next, in the state shown in FIG.
It is assumed that a failure has occurred or a LAN switching instruction has been issued by a user operation. In this case, the computers 10A and 10B connect their own first system NIC 101 that is logically incorporated into the LAN switching unit 10A.
5 from the system (step S102), and then the LAN switching unit 1
05, the shared address setting function
P and IP 'are additionally set to its own second system NIC 102 (step S103). The state of the system at this time is shown in FIG.

As described above, the switching of the LAN can be easily realized as in the sixth embodiment.

[Tenth Embodiment] Next, a tenth embodiment of the present invention will be described.
An embodiment will be described. The system configuration is
FIG. 14 is referred to for convenience.

The feature of this embodiment is that the switching instruction unit 200 shown in FIG. 14 is provided with a LAN (by applying the seventh embodiment) to another computer (here, the computer 20B) by deleting and setting a shared address. In place of the function of instructing the execution of the switching process, a function of instructing the execution of the LAN switching process by disconnecting the NIC from the system and setting the shared address is provided. Hereinafter, the LAN switching according to the present embodiment will be described with reference to the flowchart in FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 20A and 20B, the shared address setting function of the LAN switching unit 105 causes the shared addresses IP and IP. 'Is additionally set to its own first system NIC 101 (steps S111 and S121). The state of the system at this time is the same as that in FIG. If necessary, the reference numerals 10A and 10B in FIG.
A, 20B.

Then, a failure of the LAN 11 occurs in such a state, and the computer 20A detects that a failure has occurred.
Alternatively, the user operates the computer 20A to access the LAN
As a result of inputting the switching instruction, it is assumed that the computer 20A has accepted the switching instruction.

In this case, the computer 20A uses the normal LAN 12 (and the IP2 and IP2 ') to the other computer (here, the computer 20B) by the switching instruction section 200 (see the ninth embodiment). Instruct the execution of the LAN switching process (in the same manner as in the embodiment) (step S112). Computer 20A
In the same manner as in the ninth embodiment, the first NIC 101, which is logically incorporated, is separated from the system by the NIC disconnection function of the LAN switching unit 105 (step S113), and then the LAN switching is performed. The shared address setting function of the unit 105 additionally sets the shared address IP in the own second NIC 102 (step S114).

On the other hand, in the computer 20B in which the execution of the LAN switching processing is instructed by the switching instructing unit 200 in the computer 20A, the NIC 101 of its own first system, which is logically incorporated, is connected to the LAN switching unit 1
05 by the NIC disconnection function (step S122), and then the LAN switching unit 1
05, the shared address setting function
P ′ is additionally set to its own second system NIC 102 (step S123). The state of the system at this time is
This is the same as FIG. If necessary, FIG. 21 (b)
, Please replace the symbols 10A, 10B with 20A, 20B.

As described above, the switching of the LAN can be easily realized as in the seventh embodiment.

[Eleventh Embodiment] Next, an eleventh embodiment of the present invention will be described.
An embodiment will be described. The system configuration is
FIG. 17 is referred to for convenience.

The feature of the present embodiment is that the other computer (here, the computer 30B) applied to the other computer LAN switching unit 302 in FIG. In place of the function of executing the LAN switching processing by the deletion and addition setting, a function of executing the LAN switching processing by disconnecting the NIC from the system and adding the shared address by remote processing is provided. Hereinafter, the LAN switching according to the present embodiment will be described with reference to the flowchart in FIG.

First, in all the computers connected to the first system LAN 11 and the second system LAN 12, for example, the computers 30A and 30B, the shared address IP, IP 'Is additionally set to its own first system NIC 101 (steps S131 and S141). The state of the system at this time is the same as that in FIG. If necessary, reference numerals 10A and 10B in FIG.
A, 30B.

Next, a failure of the LAN 11 occurs in such a state, and the computer 30A detects that a failure has occurred, or the user operates the computer 30A to input a LAN switching instruction. It is assumed that the data is received by the computer 30A.

In this case, in the computer 30A, the normal computer LA switching unit 302
Using N12 (and IP2, IP2 '), by remote processing, another computer (here, computer 3)
0B) N of the first system logically incorporated in
Disconnect IC 101 from the system (step S13)
2). In the computer 30A, this step S1
In parallel with 32, its own logically embedded first
The NIC 101 of the system is disconnected from the system by the NIC disconnection function of the LAN switching unit 105 in the own computer (step S133).

Subsequently, in the computer 30A, the normal LAN 1
2 (and IP2, IP2 '), the shared address IP' is changed to the computer 30B which is another computer.
Are additionally set to the NIC 102 of the second system by remote processing (step S134). Computer 3
At 0A, in parallel with step S134, the LAN switching unit 105 in the own computer additionally sets the shared address IP to its own second NIC 102 (step S135). The state of the system at this time is the same as that in FIG. If necessary, Figure 21
In (b), please replace 10A and 10B with 30A and 30B.

As described above, the switching of the LAN can be easily realized similarly to the eighth embodiment.

[Twelfth Embodiment] Next, a twelfth embodiment of the present invention will be described.
An embodiment will be described. The system configuration is
FIG. 1 is referred to for convenience. However, the address storage unit 104
Stores the addresses set in the NICs 101 and 102 of the own computer.

This embodiment is different from the sixth embodiment in that the shared addresses IP and IP 'are set in the NIC 101 in a state where the NICs 101 and 102 are incorporated in the systems in the respective computers 10A and 10B. Is unnecessary. Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart in FIG. 24 and the system state diagram in FIG.

First, as shown in FIG. 25A, the NICs 101 and 102 of the computer 10A
It is assumed that P addresses IP1 and IP2 are set in advance, and IP addresses IP1 'and IP2' are set in the NICs 101 and 102 of the computer 10B, respectively. In this state, the computers 10A and 10B
Communication can be performed using the LAN 11 or the LAN 12 by IP1, IP1 'or IP2, IP2'.

In this state, it is assumed that a failure of the LAN 11 has occurred or a LAN switching instruction has been issued by a user operation. In this case, the computer 10
In steps A and 10B, the address deletion function of the LAN switching unit 105 deletes the first system addresses IP1 and IP1 'set in the own first system NIC 101 (step S151). Part 1
The first IP address IP1 and IP1 'of the first system are transferred to the NIC 102 of the second system by the address setting function of the NIC 102 of the second system.
Is additionally set (step S152). The state of the system at this time is shown in FIG.

As described above, LAN switching can be easily realized. In this state, the computer 10A,
10B indicates that the LAN to be used is from LAN11 to LAN12.
However, it is possible to communicate with each other using the same first system addresses IP1 and IP1 'as before the failure occurred.

[Thirteenth Embodiment] A thirteenth embodiment of the present invention will now be described.
An embodiment will be described. The system configuration is
FIG. 14 is referred to for convenience. However, the address storage unit 10
4 stores the addresses set in the NICs 101 and 102 of the own computer.

This embodiment is characterized in that the technical concept of the twelfth embodiment is applied to the seventh embodiment. Hereinafter, the LAN switching according to the present embodiment will be described with reference to the flowchart in FIG.

First, the NIC 10 of the computer 20A
IP addresses IP1, IP2
Are set in advance, and the NIC 101,
102 has IP addresses IP1 'and IP2', respectively.
Is set in advance. This state is shown in FIG.
Same as (a). If necessary, replace the reference numerals 10A and 10B with 20A and 20B in FIG.

In this state, a failure of the LAN 11 occurs, and the computer 20A detects that a failure has occurred, or as a result of the user operating the computer 20A and inputting a LAN switching instruction, the switching instruction is transmitted to the computer 20A. It shall be accepted in. In this case, in the computer 20A, the normal LAN 12 (and IP2, IP2 ') is used for the other computer (here, the computer 20B) by the switching instruction unit 200 (step S151 in the twelfth embodiment).
An instruction is issued to execute LAN switching processing (in the same manner as in step S152) (step S161). In the computer 20A, similarly to steps S151 and S152 in the twelfth embodiment, the LAN switching unit 105
Of its own first system NIC1
01 is deleted, and then the address setting function of the LAN switching unit 105 additionally sets the first system address IP1 to its own second system NIC 102. Processing (Step S
162 and S163) are performed.

On the other hand, in the computer 20B in which the execution of the LAN switching processing is instructed by the switching instructing unit 200 in the computer 20A, the address deletion function of the LAN switching unit 105 causes the own NIC 10 of its own system to be executed.
The first system address IP1 'set to 1 is deleted, and then the first system address IP1' is added to its own second system NIC 102 by the address setting function of the LAN switching unit 105. The setting process (steps S171 and S172) is performed. At this time,
This is the same as FIG. If necessary, FIG. 25 (b)
, Please replace the symbols 10A, 10B with 20A, 20B.

[Fourteenth Embodiment] Next, a fourteenth embodiment of the present invention will be described.
An embodiment will be described. The system configuration is
FIG. 17 is referred to for convenience. However, the address storage unit 10
4 stores the addresses set in the NICs 101 and 102 of its own computer, and the other computer address storage unit 301 stores the addresses set in the NICs 101 and 102 of the other computer (here, the computer 30B). Shall be.

This embodiment is characterized in that the technical concept of the twelfth embodiment is applied to the eighth embodiment. Hereinafter, the LAN switching in the present embodiment will be described with reference to the flowchart in FIG.

First, the NIC 10 of the computer 30A
IP addresses IP1, IP2
Are set in advance, and the NIC 101,
102 has IP addresses IP1 'and IP2', respectively.
Is set in advance. This state is shown in FIG.
Same as (a). If necessary, replace reference numerals 10A and 10B with 30A and 30B in FIG.

In this state, a failure of the LAN 11 occurs, and the computer 30A detects that a failure has occurred, or as a result of the user operating the computer 30A and inputting a LAN switching instruction, the switching instruction is transmitted to the computer 30A. It shall be accepted in.

In this case, in the computer 30A, the normal computer LA switching unit 302
Using the N12 (and IP2, IP2 '), the first system NIC 10 in the other computer (computer 30B) stored in the other computer address storage unit 301
The address of the first system set to 1, ie, IP1 ',
It is deleted by remote processing from the NIC 101 of the first system in the other computer (computer 30B) (step S181). Further, in parallel with this step S181, the computer 30A
The AN switching unit 105 controls the NIC of its own first system.
The first system address IP1 is deleted from 101 (step S182).

Subsequently, in the computer 30A, the normal LAN 1
2 (and IP2, IP2 '), the first system address IP1' is additionally set to the second system NIC 102 in the computer 30B, which is another computer, by remote processing (step S183). In parallel with this step S183, the computer 30A causes the LAN switching unit 105 in its own computer to send its first system address I to its own second system NIC 102.
P1 is additionally set (step S184). The state of the system at this time is the same as that in FIG. If necessary, reference numerals 10A and 10B are replaced by 30 in FIG.
A, 30B.

[Fifteenth Embodiment] The fifteenth embodiment of the present invention will now be described.
An embodiment will be described. The system configuration is
FIG. 1 is referred to for convenience.

The feature of this embodiment is that, as in the twelfth embodiment, each of the computers 10A and 10B has N
The point is that, in a state where the ICs 101 and 102 are incorporated in the system, the process of setting the shared addresses IP and IP 'in the NIC 101 is unnecessary. However, in this embodiment,
The difference from the twelfth embodiment is that the NIC 101 is separated from the system instead of deleting the addresses (IP1, IP1 ') set in the NIC 101 of the computers 10A and 10B. Less than,
The LAN switching according to the present embodiment will be described with reference to the flowchart in FIG. 28 and the system state diagram in FIG.

First, as shown in FIG. 29A, the NICs 101 and 102 of the computer 10A
It is assumed that P addresses IP1 and IP2 are set in advance, and IP addresses IP1 'and IP2' are set in the NICs 101 and 102 of the computer 10B, respectively. In this state, the computers 10A and 10B
The LAN is determined by the addresses IP1 and IP1 'of the first system or the addresses IP2 and IP2' of the second system.
11 or the LAN 12 can be used for communication.

In this state, it is assumed that a failure has occurred in the LAN 11 or a LAN switching instruction has been issued by a user operation. In this case, the computer 10
In A and 10B, the first of their own logically incorporated
Network NIC 101 to the LAN switching unit 105
The system is separated from the system by the C separation function (step S191), and thereafter, the addresses IP1 and IP1 'of the first system (stored in the own address storage unit 104) by the address setting function of the LAN switching unit 105.
Is additionally set to its own second system NIC 102 (step S192). FIG. 29 shows the state of the system at this time.
(B).

As described above, the switching of the LAN can be easily realized.

[Sixteenth Embodiment] A sixteenth embodiment of the present invention will now be described.
An embodiment will be described. The system configuration is
FIG. 14 is referred to for convenience.

The feature of this embodiment is that, similarly to the thirteenth embodiment, each computer 20A, 20B has N
The point is that, in a state where the ICs 101 and 102 are incorporated in the system, the process of setting the shared addresses IP and IP 'in the NIC 101 is unnecessary. However, in this embodiment,
The third embodiment differs from the thirteenth embodiment in that the NIC 101 is separated from the system instead of deleting the addresses (IP1, IP1 ') set in the NIC 101 of the computers 20A and 20B. That is, the feature of the present embodiment lies in that the technology of separating the NIC 101 from the system instead of deleting the address set in the NIC 101 is applied to the thirteenth embodiment. Hereinafter, the LAN switching according to the present embodiment will be described with reference to the flowchart in FIG.

First, the NIC 10 of the computer 20A
IP addresses IP1, IP2
Are set in advance, and the NIC 101,
102 has IP addresses IP1 'and IP2', respectively.
Is set in advance. This state is shown in FIG.
Same as (a). If necessary, replace 10A and 10B with 20A and 20B in FIG.

In this state, a failure of the LAN 11 occurs, and the computer 20A detects that a failure has occurred, or as a result of the user operating the computer 20A and inputting a LAN switching instruction, the switching instruction is transmitted to the computer 20A. It shall be accepted in. In this case, in the computer 20A, the normal LAN 12 (and IP2, IP2 ') is used for the other computer (here, the computer 20B) by the switching instruction unit 200 (step S191 in the fifteenth embodiment).
An instruction to execute LAN switching processing (in the same manner as in step S192) is issued (step S201). In the computer 20A, the NIC 101 of its own first system, which is logically incorporated, is separated from the system by the NIC disconnection function of the LAN switching unit 105 as in steps S191 and S192 in the fifteenth embodiment. Thereafter, the address setting function of the LAN switching unit 105 additionally sets the address IP1 of the first system (stored in the own address storage unit 104) to the own NIC 102 of the second system (step S202,
S203) is performed.

On the other hand, in the computer 20B in which the execution of the LAN switching processing is instructed by the switching instructing unit 200 in the computer 20A, the NIC 101 of its own first system, which is logically incorporated, is connected to the LAN switching unit 1
05, and then the address setting function of the LAN switching unit 105, and then the first system address IP1 '(stored in its own address storage unit 104) is used by its own second system. Of additional setting to the NIC 102 (step S21)
1, S212) is performed. The state at this time is shown in FIG.
Same as (b). If necessary, replace reference numerals 10A and 10B with 20A and 20B in FIG. 29 (b).

As described above, LAN switching can be easily realized.

[Seventeenth Embodiment] The seventeenth embodiment of the present invention will now be described.
An embodiment will be described. The system configuration is
FIG. 17 is referred to for convenience.

The feature of this embodiment is that the computers 30A and 30B have N
The point is that, in a state where the ICs 101 and 102 are incorporated in the system, the process of setting the shared addresses IP and IP 'in the NIC 101 is unnecessary. However, in this embodiment,
The fourth embodiment is different from the fourteenth embodiment in that the NIC 101 is separated from the system instead of deleting the addresses (IP1, IP1 ') set in the NIC 101 of the computers 30A and 30B. That is, the feature of the present embodiment lies in that the technology of disconnecting the NIC 101 from the system instead of deleting the address set in the NIC 101 is applied to the fourteenth embodiment. Hereinafter, the LAN switching according to the present embodiment will be described with reference to the flowchart in FIG.

First, the NIC 10 of the computer 30A
IP addresses IP1, IP2
Are set in advance, and the NIC 101,
102 has IP addresses IP1 'and IP2', respectively.
Is set in advance. This state is shown in FIG.
Same as (a). If necessary, replace the reference numerals 10A and 10B with 30A and 30B in FIG.

In this state, a failure of the LAN 11 occurs, and the computer 30A detects that a failure has occurred, or as a result of the user operating the computer 30A and inputting a LAN switching instruction, the switching instruction is transmitted to the computer 30A. It shall be accepted in.

In this case, in the computer 30A, the normal computer LA switching unit 302
Using N12 (and IP2, IP2 '), by remote processing, another computer (here, computer 3)
0B) N of the first system logically incorporated in
Disconnect IC 101 from the system (step S22)
1). In the computer 30A, this step S
In parallel with the H.221, the first NIC 101 that is logically incorporated is disconnected from the system by the NIC disconnection function of the LAN switching unit 105 in the own computer (step S222).

Subsequently, in the computer 30A, the normal LAN 1
2 (and IP2, IP2 '), the address of the first system applied to the computer 30B, which is another computer, stored in the other computer address storage unit 301, that is, IP1', is stored in the computer 30B. Is additionally set to the second NIC 102 by remote processing (step S223). Also, in the computer 30A, in parallel with the step S223, the LAN switching unit 105 in the own computer causes the address storage unit 10
4 is additionally set to its own second system NIC 102 (step S5).
224). The state of the system at this time is the same as that in FIG. If necessary, reference numeral 1 in FIG.
Replace 0A, 10B with 30A, 30B.

As described above, LAN switching can be easily realized.

[Eighteenth Embodiment] The eighteenth embodiment of the present invention will now be described.
An embodiment will be described.

FIG. 32 is a block diagram showing the configuration of a LAN-resistant computer system according to the eighteenth embodiment of the present invention, and the same parts as those in FIG. 1 are denoted by the same reference numerals.

In the figure, a plurality of computers, for example, computers 40A and 40B are connected to the first system LAN 11 and the second system LAN 12. The computers 40A and 40B are connected to the first LAN 11
A first system NIC 101 serving as a connection port with the
The NI of the second system which forms a connection port with the LAN 12 of the second system
C102, the input / output processor 403 connected to the NICs 101 and 102, and communication path information, for example, information on which NIC in the own computer to use when communicating with a specified address (IP address). And a routing table 404 in which entries are registered.

The input / output processor 403 has a LAN switching unit 405. The LAN switching unit 405
If LAN switching is necessary, such as when a failure occurs in the LAN being used, the network address is changed to the LAN before switching.
Route information for communication with an IP address (computer destined to the destination) that indicates (in this case, N in the own computer that forms a connection port with the LAN after switching)
It has a function of additionally registering an entry of (route information indicating that communication is performed using an IC) at the head search position of the routing table 404 (that is, the position searched first). Further, the LAN switching unit 405 has a function of deleting the corresponding additional registration entry from the routing table 404 when it is necessary to return the LAN switching to the original state, for example, when the failure is solved.

Next, the LAN switching of each computer in the system having the configuration of FIG. 32 will be described with reference to the flowcharts of FIGS. 33 and 34 and the system state diagram of FIG.

First, the NIC 10 of the computer 40A
As shown in FIG. 35A, IP addresses IP1 and IP2 are set in advance in NICs 101 and 102, respectively.
It is assumed that addresses IP1 'and IP2' are set in advance. It is also assumed that the NICs 101 and 102 are assigned network addresses IPN1 and IPN2, respectively.

In this state, for example, the computer 40A
When communication is performed from the server to the address IP1 ', the NIC 101 of the first system of the computer 40A is used. Similarly, when communication from the computer 40B to the address IP1 is performed, the first
Is used. That is, the communication between the IP1 and the IP1 'by the computers 40A and 40B is performed by the NIC of the computer 40A as shown in FIG.
101-LAN11-NIC10 of computer 40B
1 is performed.

In this state, it is assumed that a failure has occurred in the LAN 11 or a LAN switching instruction has been issued by a user operation. In this case, in the computers 40A and 40B, the LAN switching unit 405
I whose network address indicates the LAN 11 of the first system
Path information of a detour when communicating with the P addresses IP1 'and IP1, that is, IP1' and IP1 using the second system NIC 102 (connected to the second system LAN 12)
Further, an entry of the route information indicating communication with No. 1 is additionally registered at the head search position of the routing table 404 (steps S231 and S241). In this case, the search position (logical or physical position) of each registered information entry is moved down by one entry.

Then, in the computers 40A and 40B,
When communicating with the addresses IP1 'and IP1, the IP
By searching the routing table 404 with 1 'and IP1, it is possible to first obtain the route information of the detour for communicating with the IP1' and IP1 additionally registered at the first search position. Thereby, the communication between IP1 and IP1 'by the computers 40A and 40B is
As shown in FIG. 35 (c), the NI of the computer 40A
C101-NIC of computer 40A-LAN
The process is performed on the route of 12-NIC 102 of the computer 40B-NIC 101 of the computer 40B.

Here, it is assumed that the failure of the LAN 11 has been recovered, or that an instruction to restore the LAN switching has been issued by a user operation. In this case, the computer 40
A, 40B, when communicating with the entry additionally registered earlier, that is, address IP1 ', IP1, N2 of the second system
The entry of the route information of the detour indicating that communication is to be performed using the IC 102 is deleted from the routing table 404 (steps S251 and S261). In this case, the search position of each of the other registered information entries is moved up by one entry. This routing table 404
Can be restored by the operation (entry deletion operation) for.

Since the above LAN switching can be performed for each computer, for example, as shown in FIG.
In a system in which computers 40A, 40B, and 40C are connected to 11 and 12, respectively,
0A, 40B, first system address IP1, IP
1 'is communicated with the NIC 101-L of the computer 40A.
The communication between the addresses IP1 and IP1 'of the first system by the computers 40A and 40C is performed by the computer 40A via the route of the NIC 101 of the AN11 and the computer 40B.
NIC 101-NIC 102 of computer 40A
It is also possible to perform the processing through the route of LAN 12-NIC 102 of computer 40C-NIC 101 of computer 40C.

The above LAN switching is performed by the LAN 11
And 12 at the same time,
It is also possible to switch between 1 → LAN12 and LAN12 → LAN11.

In the eighteenth embodiment described above, the registration of the route information entry in the routing table 404 in each computer and the deletion of the route information entry from the routing table 404 are autonomously performed by the computer. Although the description has been given of the case in which this is performed, the present invention is not limited to this. For example, as in the seventh embodiment, one computer that has detected a failure such as a LAN,
Alternatively, the computer that has accepted the LAN switching instruction from the user instructs all other computers to execute registration / deletion of the path information entry using the normal LAN (the LAN 12 in the example of FIG. 35). Alternatively, the registration / deletion processing of the path information entry may be executed in the own computer. Similarly to the eighth embodiment, one computer that has detected a failure such as a LAN, or a computer that has received a user's instruction to switch to a LAN, determines whether a routing information entry in the routing table 404 in another computer is correct. Registration / deletion is performed on a normal LAN (LAN1 in the example of FIG. 35).
The processing may be performed by remote processing using 2).

In the eighteenth embodiment, the registration of the route information entry in the
The case where the operation is performed in P address units has been described, but the present invention is not limited to this. For example, as shown in the flowchart of FIG. 37, the registration of the route information entry may be performed in network address units. In this case, entry deletion is performed in network address units as shown in the flowchart of FIG. In this method, IP
The number of registered entries can be reduced as compared with the case where registration is performed in address units.

[Nineteenth Embodiment] Next, a nineteenth embodiment of the present invention will be described.
An embodiment will be described.

FIG. 39 is a block diagram showing a configuration of a LAN-resistant computer system according to the nineteenth embodiment of the present invention.

In the figure, a plurality of computers, for example, computers 50A and 50B are connected to the duplicated LANs (transmission paths) 51 and 52. The computer 50A is a TCP / IP server that executes a TCP / IP server program, and the computer 50B is a TCP / IP server.
TCP that runs CP / IP client programs
/ IP client.

Computers 50A and 50B are LAN
Communication is performed using 51 or 52. Here is LAN
51 and 52 are treated as equal LANs without using one as a main system and the other as a subordinate system as in the related art. Each computer 50
A and 50B are NICs serving as connection ports with the LAN 51
NIC 50 serving as a connection port between LAN 501 and LAN 52
2 and a routing table 504.

The NICs 501 and 50 of the computer 50A
2, IP addresses X1 and Y1 are set, and NICs 501 and 502 of the computer 50B are set with IP addresses X2 and Y2.

The routing table 504 has a destination (Destination) as described in the prior art.
A pair of the IP address of the destination and the IP address of the destination (inquiry destination) is registered. When the LANs 51 and 52 can communicate with each other, default contents are set in the routing table 504 as in the related art. As shown in FIG. 39, the default contents of the routing tables 504 of the computers 50A and 50B are as follows: * Computer 50A Destination = X2, Gateway = X2 Destination = Y2, Gateway = Y2 * Computer 50B Destination = X1, Gateway = X1 Destination = Y1, Gateway = Y1.

This embodiment is characterized in that the LAN 51 or 5
When communication becomes impossible (when there is an uncommunicable LAN), as shown in the flowchart of FIG.
Destination connection port (NI
One latest detour route (information entry) to the address set in C) is placed in the head search position of its own routing table 504 (a position searched before other route information entries related to the address). Is added (step S301), and all the other detour routes (information entries thereof) are deleted (steps S302, S303).
It is in.

Therefore, for example, when the LAN 52 cannot communicate, in the computer 50A, Destination = Y2 and Gateway = X2 are added. If there is another detour route Destination = X2, Gateway = Y2, it is deleted. This detour route is added when the LAN 51 cannot communicate as described later.

On the other hand, in the computer 50B, Destination = Y1 and Gateway = X1 are added. Further, if there is another detour route Destination = X1, Gateway = Y1, it is deleted. This detour route is added when the LAN 51 cannot communicate as described later.

Similarly, when the LAN 51 cannot communicate, in the computer 50A, Destination = X2 and Gateway = Y2 are added. If there is another detour route Destination = Y2, Gateway = X2, it is deleted.

On the other hand, in the computer 50B, Destination = X1 and Gateway = Y1 are added. Also, if there is another detour route Destination = Y1, Gateway = X1, it is deleted.

Next, the connection between the computers 50A and 50B when the LAN 51 can communicate and the connection between the computers 50A and 50B when the LAN 51 cannot communicate in the system of FIG. 41 and 42 will be sequentially described.

(1) Connection when LAN 51 is communicable It is assumed that LAN 51 is in a communicable state. At this time, it is assumed that the routing tables 504 of the computers 50A and 50B have default contents, for example, as shown in FIG. In this state, when a connection is made from the computer (client) 50B to the computer (server) 50A (X1), the address X2 (as a source address) of the NIC 101 on the LAN 51 side that can communicate without explicitly binding a socket. Automatically bound. That is,
X2 is the TCP / IP source address.

In such a state, when the LAN 51 becomes unable to communicate, Destination = X2 and Gateway = Y2 are added to the routing table 504 of the computer 50A as shown in FIG. 41B (step S301). At this time, if there is another detour route Destination = Y2, Gateway = X2, they are deleted (steps S302, S3).
03).

Similarly, as shown in FIG. 41B, Destination = X1 and Gateway = Y1 are added to the routing table 504 of the computer 50B (step S301). At this time, if there is another detour route Destination = Y1, Gateway = X1, they are deleted (steps S302, S3).
03).

In this case, the computer (client)
50B intends to transmit to X1, but can transmit a request using LAN 52 because Destination = X1, Gateway = Y1 exists in its own routing table 504.
Therefore, communication can be continued while maintaining a TCP / IP session.

On the other hand, the computer (server) 50A
A response is to be sent to the TCP / IP source address X2. However, since the own routing table 504 has Destination = X2 and Gateway = Y2, a response can be returned using the LAN 52. Therefore, communication can be continued while maintaining a TCP / IP session.

(2) Connection when LAN 51 cannot communicate First LAN 51 cannot communicate and LAN 52 can communicate
(Client) 50B in the state of
Is connected to the computer (server) 50A (of X1), it is automatically bound (as a source address) to the address Y2 of the NIC 102 on the side of the communicable LAN 52 without explicitly binding the socket. That is, Y2 is the source address of TCP / IP.

When the LAN 52 enters the second state in which the LAN 52 cannot communicate and the LAN 51 can communicate in the first state, in the present embodiment where the LANs 51 and 52 are equal, the routing tables 504 of the computers 50A and 50B are provided.
Is not the default content unlike the past, and is as follows.

First, in the computer 50A, as shown in FIG. 42, the NIC on the LAN 52 side of the computer 50B
One detour route to Y2 set to 502, Destination = Y2, Gateway = X2, is added (step S301), and the other preset detour routes, Destination = X2, Gateway = Y2, are deleted (step S301). S302, S303).

On the other hand, in the computer 50B, as shown in FIG. 42, one detour route Destination = Y1, Gateway = X1 to Y1 set in the NIC 502 on the LAN 52 side of the computer 50A is added (step S301). Then, the other preset detour routes Destination = X1, Gateway = Y1 are deleted (steps S302, S303).

In the state of FIG. 42, the computer (client) 50B tries to transmit to X1, but default routing information Destination = X1, Gateway is stored in its own routing table 504 (the position searched first with respect to X1). = Y1, the request can be transmitted using the LAN 52.
Therefore, communication can be continued while maintaining a TCP / IP session.

On the other hand, the computer (server) 50A
Attempts to return a response to the source address Y2 of TCP / IP, but there is Destination = Y2, Gateway = X2 at the head search position (the position searched first for Y2) of its own routing table 504, so that it is later than that. Even if there is default route information Destination = Y2, Gateway = Y2 destined for the same Y2, a response can be returned using the LAN 51 without being affected by it. Therefore, TCP /
Communication can be continued while maintaining the IP session.

[Twentieth Embodiment] A connection between computers using LAN switching by a routing table operation according to the flowchart of FIG. 40 applied in the nineteenth embodiment is a LAN-resistant LAN-tolerant fault. The present invention is not limited to the computer system, and can be applied to a LAN-resistant computer system having a triple LAN.

Therefore, a LAN switching method by operating the routing table according to the flowchart of FIG.
A twentieth embodiment of the present invention applied to a LAN-resistant computer system with a triple LAN is described.

FIG. 43 is a block diagram showing the configuration of a LAN-resistant computer system according to the twentieth embodiment of the present invention, and the same parts as those in FIG. 39 are denoted by the same reference numerals.

In the figure, a triple LAN (3-system LA
N), three LANs 51, 52, and 53 include a plurality of computers, for example, computers 60A and 60A.
B is connected. Computer 60A is TCP / I
The computer 60B is a TCP / IP client that executes a TCP / IP client program. Here, the LANs 51, 52, and 53 are treated as equivalent LANs.

Each of the computers 60A and 60B is connected to a LAN.
An NIC 501 serving as a connection port with the LAN 51 and a LAN 52
An NIC 502 serving as a connection port with the LAN 53, an NIC 503 serving as a connection port with the LAN 53, and a routing table 604 are provided.

The NICs 501 and 50 of the computer 60A
2, 503, IP addresses X1, Y1, Z1 are set, and NICs 501, 502, 5 of the computer 60B are set.
03 is assumed to have IP addresses X2, Y2, and Z2 set therein.

[0199] In the routing table 604, a pair of an IP address of a destination (Destination) and an IP address of a destination (inquiry destination) is registered. L
If all of the ANs 51, 52, 53 can communicate,
Default contents are set in the routing table 604. The default contents of the routing table 604 of the computers 60A and 60B are shown in FIG.
* Computer 60A Destination = X2, Gateway = X2 Destination = Y2, Gateway = Y2 Destination = Z2, Gateway = Z2 * Computer 60B Destination = X1, Gateway = X1 Destination = Y1, Gateway1Y = Y1ZayYay = Y1 Gateway = Z1.

This embodiment is characterized in that, as in the case of the nineteenth embodiment, when any of the LANs 51, 52, and 53 becomes unable to communicate (when there is a LAN that cannot communicate), FIG. As shown in the flowchart of FIG. 7, one detour route (information entry) to the address set to the destination connection port (NIC) on the LAN side where communication is disabled is located at the head search position of its own routing table 604. (A position searched prior to another route information entry related to the address) (step S30).
1) Deleting all (information entries of) the other detour routes (steps S302 and S303).

The following describes the routing table operation in the present embodiment when the LAN 51 cannot communicate, when the LAN 52 cannot communicate, and when the LAN 53 cannot communicate.

(1) When the LAN 51 cannot communicate (1-1) When the LAN 52 can communicate When the LAN 51 cannot communicate, if the LAN 52 can communicate (regardless of the state of the LAN 53), the computer 60A , 60B are as follows by operating the routing table according to the flowchart of FIG.

First, in the computer 60A, one bypass route Destination = X2, Gateway = Y2 to X2 set in the NIC 501 on the LAN 51 side of the computer 60B is added (step S301). Also, if there are other preset detour routes Destination = X2, Gateway = Z2, Destination = Y2, Gateway = X2, Destination = Z2, and Gateway = X2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to X2 and the default contents.

On the other hand, in the computer 60B, one detour route Destination = X1, Gateway = Y1 to X1 set in the NIC 501 on the LAN 51 side of the computer 60A is added (step S301). Also, if there are other preset detour routes Destination = X1, Gateway = Z1 Destination = Y1, Gateway = X1 Destination = Z1, Gateway = X1, if they exist, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to X1 and the default contents.

(1-2) When the LAN 53 Can Communicate When the LAN 51 cannot communicate, and when the LAN 53 can communicate (regardless of the state of the LAN 52), the contents of the routing tables 604 of the computers 60A and 60B are as follows. The following is the result of the routing table operation according to the flowchart of FIG.

First, in the computer 60A, one bypass route Destination = X2, Gateway = Z2 to X2 set in the NIC 501 on the LAN 51 side of the computer 60B is added (step S301). Also, if there are other preset detour routes Destination = X2, Gateway = Y2, Destination = Y2, Gateway = X2, Destination = Z2, and Gateway = X2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to X2 and the default contents.

On the other hand, in the computer 60B, one bypass route Destination = X1 and Gateway = Z1 to X1 set in the NIC 501 on the LAN 51 side of the computer 60A is added (step S301). Also, if there are other preset detour routes Destination = X1, Gateway = Y1 Destination = Y1, Gateway = X1 Destination = Z1, Gateway = X1 if they exist, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to X1 and the default contents.

(2) When the LAN 52 cannot communicate (2-1) When the LAN 51 can communicate When the LAN 52 cannot communicate, if the LAN 51 can communicate (regardless of the state of the LAN 53), the computer 60A , 60B are as follows by operating the routing table according to the flowchart of FIG.

First, in the computer 60A, one bypass route Destination = Y2, Gateway = X2 to Y2 set in the NIC 502 on the LAN 52 side of the computer 60B is added (step S301). Also, if there are other preset detour routes Destination = Y2, Gateway = Z2 Destination = X2, Gateway = Y2 Destination = Z2, Gateway = Y2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to Y2 and the default contents.

On the other hand, in the computer 60B, one bypass route Destination = Y1, Gateway = X1 to Y1 set in the NIC 502 on the LAN 52 side of the computer 60A is added (step S301). Also, if there are other preset detour routes Destination = Y1, Gateway = Z1 Destination = X1, Gateway = Y1 Destination = Z1, Gateway = Y1, if there are any, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to Y1 and the default contents.

(2-2) When the LAN 53 Can Communicate When the LAN 52 cannot communicate, and when the LAN 53 can communicate (regardless of the state of the LAN 51), the contents of the routing tables 604 of the computers 60A and 60B are as follows. The following is the result of the routing table operation according to the flowchart of FIG.

First, in the computer 60A, one detour route Destination = Y2, Gateway = Z2 to Y2 set in the NIC 502 on the LAN 52 side of the computer 60B is added (step S301). If there are other preset detour routes Destination = Y2, Gateway = X2, Destination = X2, Gateway = Y2, Destination = Z2, and Gateway = Y2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to Y2 and the default contents.

On the other hand, in the computer 60B, one bypass route Destination = Y1, Gateway = Z1 to Y1 set in the NIC 502 on the LAN 52 side of the computer 60A is added (step S301). Also, if there are other preset detour routes Destination = Y1, Gateway = X1 Destination = X1, Gateway = Y1 Destination = Z1, Gateway = Y1, if there are any, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to Y1 and the default contents.

(3) When the LAN 53 cannot communicate (3-1) When the LAN 51 can communicate When the LAN 53 cannot communicate, if the LAN 51 can communicate (regardless of the state of the LAN 52), the computer 60A , 60B are as follows by operating the routing table according to the flowchart of FIG.

First, in the computer 60A, one bypass route Destination = Z2, Gateway = X2 to Z2 set in the NIC 503 on the LAN 53 side of the computer 60B is added (step S301). If there are other preset detour routes Destination = Z2, Gateway = Y2, Destination = Y2, Gateway = Z2, Destination = X2, and Gateway = Z2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to Z2 and the default contents.

On the other hand, in the computer 60B, one detour route Destination = Z1, Gateway = X1 to Z1 set in the NIC 503 on the LAN 53 side of the computer 60A is added (step S301). In addition, if there are other preset detour routes Destination = Z1, Gateway = Y1 Destination = Y1, Gateway = Z1 Destination = X1, Gateway = Z1 if they exist, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to Z1 and the default contents.

(3-2) When the LAN 52 Can Communicate When the LAN 53 cannot communicate, and when the LAN 52 can communicate (regardless of the state of the LAN 51), the contents of the routing tables 604 of the computers 60A and 60B are as follows. The following is the result of the routing table operation according to the flowchart of FIG.

First, in the computer 60A, one bypass route Destination = Z2, Gateway = Y2 to Z2 set in the NIC 503 on the LAN 53 side of the computer 60B is added (step S301). Also, if there are other preset detour routes Destination = Z2, Gateway = X2, Destination = Y2, Gateway = Z2, Destination = X2, and Gateway = Z2, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60A are one detour route to Z2 and the default contents.

On the other hand, in the computer 60B, one detour route Destination = Z1, Gateway = Y1 to Z1 set in the NIC 503 on the LAN 53 side of the computer 60A is added (step S301). Also, if there are other preset detour routes Destination = Z1, Gateway = X1 Destination = Y1, Gateway = Z1 Destination = X1, Gateway = Z1 if they exist, they are deleted (steps S302, S3).
03).

That is, the contents of the routing table 604 of the computer 60B are one detour route to Z1 and the default contents.

[0232]

As described in detail above, according to the present invention, communication can be continued without changing the destination address used before the switching even after the network switching, and the network switching can be performed as in the prior art. Therefore, there is no need to change the destination address, thereby eliminating the need for a special communication application.

Further, according to the present invention, even if any one of the multiplexed networks becomes incapable of communication, communication can always be continued as long as a communicable network exists.

[Brief description of the drawings]

FIG. 1 is an exemplary block diagram showing a configuration of a LAN-resistant computer system according to a first embodiment of the present invention;

FIG. 2 is a flowchart for explaining the operation in the first embodiment.

FIG. 3 is a system state diagram for explaining an operation in the first embodiment.

FIG. 4 is a flowchart illustrating an operation according to the second embodiment of the present invention.

FIG. 5 is a system state diagram for explaining an operation in the second embodiment.

FIG. 6 is a flowchart for explaining an operation according to the third embodiment of the present invention.

FIG. 7 is a system state diagram for explaining an operation in the third embodiment.

FIG. 8 is a flowchart illustrating an operation according to a fourth embodiment of the present invention.

FIG. 9 is a system state diagram for explaining an operation in the fourth embodiment.

FIG. 10 is a flowchart illustrating an operation according to a fifth embodiment of the present invention.

FIG. 11 is a system state diagram for explaining the operation in the fifth embodiment.

FIG. 12 is a flowchart illustrating an operation according to a sixth embodiment of the present invention.

FIG. 13 is a system state diagram for explaining an operation in the sixth embodiment.

FIG. 14 is a block diagram showing a configuration of a LAN-resistant computer system according to a seventh embodiment of the present invention;

FIG. 15 is a flowchart for explaining the operation in the seventh embodiment.

FIG. 16 is a system state diagram for explaining the operation in the seventh embodiment.

FIG. 17 is a block diagram showing the configuration of a LAN-resistant computer system according to an eighth embodiment of the present invention;

FIG. 18 is a flowchart for explaining the operation in the eighth embodiment.

FIG. 19 is a system state diagram for explaining the operation in the eighth embodiment.

FIG. 20 is a flowchart illustrating an operation according to the ninth embodiment of the present invention.

FIG. 21 is a system state diagram for explaining the operation in the ninth embodiment.

FIG. 22 is a flowchart illustrating an operation according to the tenth embodiment of the present invention.

FIG. 23 is a flowchart illustrating an operation according to the eleventh embodiment of the present invention.

FIG. 24 is a flowchart illustrating an operation according to a twelfth embodiment of the present invention.

FIG. 25 is a system state diagram for explaining the operation in the twelfth embodiment.

FIG. 26 is a flowchart illustrating an operation according to a thirteenth embodiment of the present invention.

FIG. 27 is a flowchart illustrating an operation according to the fourteenth embodiment of the present invention.

FIG. 28 is a flowchart illustrating an operation according to the fifteenth embodiment of the present invention.

FIG. 29 is a system state diagram for explaining the operation in the fifteenth embodiment.

FIG. 30 is a flowchart illustrating an operation according to the sixteenth embodiment of the present invention.

FIG. 31 is a flowchart for explaining the operation in the seventeenth embodiment of the present invention.

FIG. 32 is a block diagram showing a configuration of a LAN-resistant computer system according to an eighteenth embodiment of the present invention;

FIG. 33 is a flowchart for explaining the operation in the eighteenth embodiment.

FIG. 34 is a flowchart for explaining the operation in the eighteenth embodiment;

FIG. 35 is a system state diagram for explaining the operation in the eighteenth embodiment.

FIG. 36 is a system state diagram for explaining a modification of the eighteenth embodiment.

FIG. 37 is a flowchart showing a modification of the eighteenth embodiment.

FIG. 38 is a flowchart showing a modification of the eighteenth embodiment.

FIG. 39 is a block diagram showing a configuration of a LAN-resistant computer system according to a nineteenth embodiment of the present invention;

FIG. 40 is a flowchart for explaining the operation in the nineteenth embodiment;

FIG. 41 is a system state diagram for explaining the operation in the nineteenth embodiment.

FIG. 42 is a system state diagram for explaining the operation in the nineteenth embodiment.

FIG. 43 is a block diagram showing a configuration of a LAN-resistant computer system according to a twentieth embodiment of the present invention;

FIG. 44 is a block diagram showing a configuration of a general LAN-resistant computer system.

[Explanation of symbols]

10A, 10B, 20A, 20B, 30A, 30B, 4
0A, 40B, 40C, 50A, 50B, 60A, 60
B: Computer 11, 51: LAN (network) of the first system 12, 52: LAN (network) of the second system 53: LAN (network) of the third system 101, 102, 501, 502, 503 ... Network interface (NIC) 103 Input / output processor 104 Address storage unit 105, 405 LAN switching unit 200 Switching instruction unit 301 Other computer address storage unit 302 Other computer LAN switching unit 404, 504 Routing table

Claims (24)

[Claims] 1. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. A network switching method applied to a system, wherein each of the computers has a network interface of the first system. By additionally setting a shared address unique to the own computer and common to the networks of the first system and the second system, communication between the computers is performed using the shared address. When a failure occurs in the network of the first system, or when there is an instruction to switch the network by the user, the computer deletes the shared address from the network interface of the first system, and deletes the shared address. A network switching method by performing a process of additionally setting a network interface to the second system network interface. 2. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. A network switching method applied to a system, wherein each of the computers has a network interface of the first system. By additionally setting a shared address unique to the own computer and common to the networks of the first system and the second system, communication between the computers is performed using the shared address. When a failure occurs in the network of the first system, or when there is an instruction for network switching by a user, the computer that has detected the failure or has been given the instruction for network switching. Executing a process of deleting the shared address from the network interface of the first system of the other computer and additionally setting the shared address on the network interface of the second system of the other computer. , While instructing using the network of the second system, The network is switched by deleting the shared address from the first network interface of the computer and additionally setting the shared address to the second network interface of the computer. A network switching method, characterized in that: 3. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network, and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. A network switching method applied to a system, wherein each of the computers has a network interface of the first system. By additionally setting a shared address unique to the own computer and common to the networks of the first system and the second system, communication between the computers is performed using the shared address. When a failure occurs in the network of the first system, or when there is an instruction for network switching by a user, the computer that has detected the failure or has been given the instruction for network switching. Removing the shared address from the network interface of the first system by remote processing using the network of the second system, and replacing the shared address with the network interface of the second system of the other computer. Additional setting by remote processing using the second system network On the other hand, by executing a process of deleting the shared address from the first system network interface of the own computer and additionally setting the shared address to the second system network interface of the own computer, A network switching method, comprising: 4. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A first network interface logically and logically incorporated into the system;
A plurality of computers each having a connection port with the network of the second system, having a second system network interface in which an address of the second system is set and which is logically incorporated in the system; A network system capable of performing communication between the respective computers using a corresponding first system or second system network according to an address set in a network interface of a first system or a second system. In the network switching method applied to the above, in each of the computers, a shared address unique to the own computer and common to the networks of the first system and the second system is additionally set in the network interface of the first system. In this way, communication between computers is performed using the shared address, In the event that a failure has occurred in the network of the first system, or when there is an instruction to switch the network by a user, the network interface of the first system logically incorporated in the system is set in each of the computers. By disconnecting from the system and executing a process of additionally setting a shared address unique to the computer and common to the first and second networks on the second system network interface A network switching method, comprising: 5. A computer connected to a network of a first system and a network of a second system, wherein the computer forms a connection port with the network of the first system, and has an address of the first system. A first network interface logically and logically incorporated into the system;
A plurality of computers each having a connection port with the network of the second system, having a second system network interface in which an address of the second system is set and which is logically incorporated in the system; A network system capable of performing communication between the respective computers using a corresponding first system or second system network according to an address set in a network interface of a first system or a second system. In the network switching method applied to the above, in each of the computers, a shared address unique to the own computer and common to the networks of the first system and the second system is additionally set in the network interface of the first system. In this way, communication between computers is performed using the shared address, In the case where a failure occurs in the network of the first system, or when there is an instruction to switch the network by the user, the failure is detected, or the computer to which the instruction to switch the network is given, For the computer, the network interface of the first system, which is logically incorporated in the system of the other computer, is disconnected from the system, and the own computer is connected to the network interface of the second system of the other computer. While using the second system network to instruct the execution of the process of additionally setting a common shared address unique to the first system and the first system network and the second system network. Network interface of the first system incorporated in the system The over scan with disconnected from the system, specific to the own computer to the network interface of the second system and the first
A network switching method characterized by performing a process of additionally setting a common shared address in a network of a second system and a network of a second system, thereby performing network switching. 6. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A first network interface logically and logically incorporated into the system;
A plurality of computers each having a connection port with the network of the second system, having a second system network interface in which an address of the second system is set and which is logically incorporated in the system; A network system capable of performing communication between the respective computers using a corresponding first system or second system network according to an address set in a network interface of a first system or a second system. In the network switching method applied to the above, in each of the computers, a shared address unique to the own computer and common to the networks of the first system and the second system is additionally set in the network interface of the first system. In this way, communication between computers is performed using the shared address, In the case where a failure occurs in the network of the first system, or when there is an instruction to switch the network by the user, the failure is detected, or the computer to which the instruction to switch the network is given, The network interface of the first system, which is logically incorporated in the system of the computer, is separated from the system by remote processing using the network of the second system. A method of additionally setting a shared address common to the networks of the first system and the second system to the network interface of the second system of the other computer by remote processing using the network of the second system. The first system logically incorporated in the system of its own computer The network interface with disconnected from the system, the second system of the network interface to the host computer with a unique and the first
A network switching method characterized by performing a process of additionally setting a common shared address in a network of a second system and a network of a second system, thereby performing network switching. 7. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. In the network switching method applied to the system, when a failure occurs in the network of the first system, or When there is an instruction to switch the network by the user, each computer deletes the address of the first system from the network interface of the first system and replaces the address of the first system with the second system. A network switching method characterized by performing network switching by executing processing for additionally setting a network interface of a system. 8. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. In the network switching method applied to the system, when a failure occurs in the network of the first system, or In the case where there is an instruction to switch the network by the user, the computer that has detected the failure or has been given the instruction to switch the network sends the first system of the other computer to all other computers. Executing the process of deleting the address of the first system from the network interface of the other computer and additionally setting the address of the first system to the network interface of the second system of the other computer;
While instructing using the network of the second system, while deleting the address of the first system from the network interface of the first system of the own computer,
A network switching method characterized by performing a process of additionally setting the address of the first system to the network interface of the second system of the own computer, thereby performing network switching. 9. A computer connected to a first system network and a second system network, wherein the computer forms a connection port with the first system network and has a first system address set therein. A plurality of computers each having a first system network interface and a second system network interface that forms a connection port with the second system network and has a second system address set therein, A network capable of performing communication between the respective computers by using an address set in a network interface of a first system or a second system using a corresponding network of the first system or the second system. In the network switching method applied to the system, when a failure occurs in the network of the first system, or When there is an instruction to switch the network by the user, the computer that has detected the failure or that has been given the instruction to switch the network uses the first system network interface of all other computers to output the first Address of the second system
The remote computer deletes the address of the first computer by using the network of the second computer and the network interface of the second computer of the other computer using the network of the second computer. While the additional setting is performed by the processing, the address of the first system is deleted from the network interface of the first system of the own computer, and
A network switching method characterized by performing a process of additionally setting the address of the first system to the network interface of the second system of the own computer, thereby performing network switching. 10. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The first system network interface in which the address of the second system is set, and the network interface of the second system which forms a connection port with the network of the second system and in which the address of the second system is set. Communication between the computers using a corresponding one of the first or second network by an address set in the network interface of the first or second system. In a network switching method applied to a network system capable of performing the following, when a failure occurs in the network of the first system, or when there is an instruction to switch the network by a user, each of the computers performs logical switching. Network interface of the first system incorporated in the system Is disconnected from the system, and the first system set to the network interface of the first system is
Switching the network by executing a process of additionally setting the address of the second system to the network interface of the second system. 11. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The first system network interface in which the address of the second system is set, and the network interface of the second system which forms a connection port with the network of the second system and in which the address of the second system is set. Communication between the computers using a corresponding one of the first or second network by an address set in the network interface of the first or second system. In a network switching method applied to a network system capable of performing, when a failure occurs in the network of the first system, or when there is an instruction for network switching by a user, the failure is detected, or On the computer given the network switching instruction, all other For the computer, the first system network interface logically incorporated in the system of the other computer is disconnected from the system, and the first system interface set in the first system network interface is disconnected. While using the second system network to instruct execution of processing for additionally setting the system address to the second system network interface of the other computer, the own computer is logically incorporated into the system. Disconnecting the first system network interface from the system, and replacing the first system address set in the first system network interface with the second system network interface of its own computer. To execute the process to add to the Ri,
A network switching method characterized by performing network switching. 12. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The first system network interface in which the address of the second system is set, and the network interface of the second system which forms a connection port with the network of the second system and in which the address of the second system is set. Communication between the computers using a corresponding one of the first or second network by an address set in the network interface of the first or second system. In a network switching method applied to a network system capable of performing, when a failure occurs in the first system network, or when there is an instruction for network switching by a user, the failure is detected, or On the computer given the network switching instruction, all other The first system network interface logically incorporated in the system of the computer is separated from the system by remote processing using the second system network, and is set as the first system network interface. The address of the first system, which has been set, is additionally set to the network interface of the second system of the other computer by remote processing using the network of the second system. And disconnects the first system network interface incorporated in the system from the system, and replaces the first system address set in the first system network interface with the second system address of its own computer. Processing to add to the network interface of the host By executing,
A network switching method characterized by performing network switching. 13. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The first system network interface in which the address of the second system is set, and the network interface of the second system which forms a connection port with the network of the second system and in which the address of the second system is set. Communication with another computer using the corresponding first or second network using an address set in the first or second network interface. An address storage means for storing a shared address unique to the own computer and common to the networks of the first system and the second system; and adding the shared address to a network interface of the first system. Setting, a first shared address setting method that enables communication between computers using the shared address. A shared address deletion unit that deletes the shared address from the network interface of the first system when a failure occurs in the network of the first system or when a user issues a network switching instruction; The shared address obtained from the second
To add to the network interface of the second system
And a network switching means having a shared address setting means. 14. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The address of the second system is set and forms a connection port between the network interface of the first system logically incorporated in the system and the network of the second system, and the address of the second system is set. And a second system network interface logically incorporated into the system together with the first system or the second system corresponding to the address set in the first system or the second system network interface. A computer capable of communicating with another computer using the network of the second system, the address storing a shared address unique to the own computer and common to the networks of the first system and the second system. Storage means, and the shared address is additionally set in the network interface of the first system. First shared address setting means for enabling communication between computers by an address, logically incorporated in the system when a failure occurs in the network of the first system or when a user issues an instruction to switch networks Disconnecting means for disconnecting the network interface of the first system from the system, and additionally setting the shared address set in the network interface of the disconnected first system to the network interface of the second system. And a network switching unit having a shared address setting unit. 15. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The address of the second system is set and forms a connection port between the network interface of the first system logically incorporated in the system and the network of the second system, and the address of the second system is set. And a second system network interface logically incorporated into the system together with the first system or the second system corresponding to the address set in the first system or the second system network interface. An address storage means for storing addresses set in the first and second network interfaces of the own computer in a computer capable of performing communication with another computer using the second system network; When a failure occurs in the network of the first system, or when the network is disconnected by a user. Address deleting means for deleting an address set in the network interface of the first system when there is an instruction of the first system, and an address set in the network interface of the first system in the second system. A network switching unit having an address setting unit for additionally setting the network interface. 16. A computer connected to a first system network and a second system network, the computer comprising a connection port with the first system network,
The address of the second system is set and forms a connection port between the network interface of the first system logically incorporated in the system and the network of the second system, and the address of the second system is set. And a second system network interface logically incorporated into the system together with the first system or the second system corresponding to the address set in the first system or the second system network interface. An address storage means for storing addresses set in the first and second network interfaces of the own computer in a computer capable of performing communication with another computer using the second system network; When a failure occurs in the network of the first system, or when the network is disconnected by a user. When there is an instruction, the disconnecting means for disconnecting the first system network interface logically incorporated in the system from the system and the address set in the first system network interface A computer comprising: a network switching unit having an address setting unit for additionally setting a second system network interface. 17. A computer connected to a first system network and a second system network, wherein the first system network interface forms a connection port with the first system network; Which one of the first and second network interfaces of the own computer to use when communicating with a designated address, and a second system network interface forming a connection port with the second system network A plurality of computers each having a routing table in which an information entry is registered, the network system being capable of performing communication among the computers using the first or second network. A network switching method, wherein the network of the first system is used. When a failure occurs during communication, or when a user issues an instruction to switch networks, each of the computers communicates with the address of the network of the first system by the network interface of the second system. Switching the network by adding an information entry indicating that the information is used in the routing table to a position searched earlier than other information entries related to communication with the address of the network of the first system. A network switching method characterized by the above-mentioned. 18. The method according to claim 18, further comprising: deleting the added information entry from the routing table when the failure is resolved after the network switching by adding the information entry or when there is a user's instruction to restore the network switching. The network switching method according to claim 17, wherein: 19. The network switching method according to claim 18, wherein the network switching by adding or deleting the information entry is performed simultaneously for the networks of the first system and the second system. 20. A computer connected to a first system network and a second system network, the first system network interface forming a connection port with the first system network; Which one of the first and second network interfaces of the own computer to use when communicating with a designated address, and a second system network interface forming a connection port with the second system network A plurality of computers each having a routing table in which an information entry is registered, the network system being capable of performing communication among the computers using the first or second network. A network switching method, wherein the network of the first system is used. If a failure occurs during communication, or if there is an instruction to switch the network by the user, the failure is detected, or the computer given the instruction to switch the network, all other computers On the other hand, an information entry indicating that the network interface of the second system is used for communication with the address of the network of the first system is set in the routing table of the computer,
While using the network of the second system to instruct execution of a process of adding to a position searched before other information entry relating to communication with the address of the network of the first system, The information entry indicating that the network interface of the second system is used for communication with the address of the network of the second system is stored in the routing table of the own computer in association with the communication with the address of the network of the first system. A network switching method characterized in that the network is switched by adding the information to a position searched before the information entry of (1). 21. A computer connected to a first system network and a second system network, wherein the first system network interface forms a connection port with the first system network; Which one of the first and second network interfaces of the own computer to use when communicating with a designated address, and a second system network interface forming a connection port with the second system network A plurality of computers each having a routing table in which an information entry is registered, the network system being capable of performing communication among the computers using the first or second network. A network switching method, wherein the network of the first system is used. If a failure occurs during communication, or if there is an instruction to switch the network by the user, the failure is detected, or the computer to which the instruction to switch the network is given. An information entry indicating that the second system network interface is used for communication with the network address is stored in the routing table of all the other computers in relation to communication with the first system network address. Is added by remote processing using the network of the second system to a position searched prior to the information entry of the network of the second system while communication with the address of the network of the first system is performed. An information entry indicating that the interface is to be used is stored in the root The grayed table, by adding the position to be searched before other information entry regarding the communication of the address of the first system network, a network switching method and performing switching of the network. 22. A computer connected to a first system network and a second system network, wherein the first system network interface forms a connection port with the first system network; Which one of the first and second network interfaces of the own computer to use when communicating with a designated address, and a second system network interface forming a connection port with the second system network And a routing table in which an information entry indicating the information entry is registered. The first system or the second system is connected to a network interface indicated by the information entry which can be searched first by searching the routing table by a target destination address. Can communicate with other computers using the same network In a computer, when a failure occurs while communicating using the first system network, or when a user issues an instruction to switch the network, the address of the first system network is An information entry indicating that the second system network interface is used for communication is placed in the routing table at a position searched before other information entries related to communication with the address of the first system network. Means for adding entries,
And a network having entry deletion means for deleting the added information entry from the routing table when the failure is resolved after the network switching by the addition of the information entry or when there is an instruction to restore the network switching by the user. A computer comprising switching means. 23. A computer connected to a multiplexed N-system network, wherein the N network interfaces form connection ports with the respective N-system networks and have respective system addresses set therein. A plurality of computers each having a routing table in which an information entry indicating a route on a communication destination computer side to a designated address is registered, and using any one of the N-system networks, In a network switching method applied to a network system capable of performing communication, the N-type network is treated equally, and if there is a network that cannot communicate, a network address indicating the network that cannot communicate Up-to-date information indicating a detour to the address with the part By adding one entry to the routing table at a position searched before the information entry representing the standard route to the address, and deleting any information entries representing the other detour routes. A network switching method characterized by performing network switching. 24. A computer connected to a multiplexed N-system network, wherein the N network interfaces form connection ports for the respective N-system networks and have respective system addresses set therein. And a routing table in which an information entry indicating a route on the communication destination computer side to the designated address is registered, and performing communication with each computer using any of the N-system networks. In a computer capable of, when there is a non-communicable network among the N-system networks, the latest information entry representing a detour path to an address having a network address portion indicating the non-communicable network, An information entry in the routing table representing a standard route to the address Computer, characterized in that it comprises a routing table operating means to add one to a position to be searched Ri destination, deleting all if any information entry representing the other detour path.