JP2000222246A - Network monitoring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a network monitoring method capable of surely judging which of plural computers constituting a cluster device executes an application program. SOLUTION: An address is previously assigned to each of the plural computers 1, 2 constituting the cluster device 3 connected to a network 4 and each application program 1b, 2b. Each of the plural computers 1, 2 issues first information showing the assigned address of the application program and that of the computer itself to a monitoring device 5 at the time of starting the execution of the program 1B, 2B. The device 5 obtains the assigned address of the application program and that of the computer itself according to the first information to judge the computer executing the application program.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a network monitoring method for monitoring a cluster device connected to a network.

[0002]

2. Description of the Related Art It is common practice to monitor whether computers such as personal computers and workstations provided on a network such as a LAN and a WAN are operating normally. The TCP / IP protocol is used for such monitoring, and the operation of each computer is determined by detecting an IP address assigned to each computer.

[0003]

However, when monitoring a cluster device that includes a plurality of computers connected to a network and executes an application program by any one of the plurality of computers, it is necessary to monitor the cluster device among the plurality of computers. It was not possible to reliably determine which one of the computers is running the application program.

It is an object of the present invention to provide a network monitoring method capable of reliably determining which of a plurality of computers constituting a cluster device is executing an application program. .

[0005]

A network monitoring method according to the present invention comprises a plurality of computers connected to a network, and one of the plurality of computers executes an application program predetermined for each of the plurality of computers. A network monitoring method for monitoring a cluster device to be monitored by a monitoring device, wherein an address is assigned to each of a plurality of computers and each application program in advance, and each of the plurality of computers assigns an application program when the application program starts executing. The monitoring device issues a first notification indicating the address and the assigned address of the computer to the monitoring device, and the monitoring device responds to the first notification to allocate the application program and the assigned address of the computer. Obtained is characterized by determining the application program execution computer.

[0006]

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

FIG. 1 shows a network system to which the network monitoring method of the present invention is applied. The computers 1 and 2 form a cluster device 3 and perform a cluster operation with each other. Each of the computers 1 and 2 includes cluster programs 1A and 2A and application programs 1B and 2B. Cluster program 1
A and 2A are the same as each other,
2 is for managing the cluster operation. The application programs 1B and 2B are the same as each other, for example, for performing a measurement operation and an automatic control operation.

The computers 1 and 2 are connected to a network cable 4. A computer 5 is further connected to the network cable 4. The computer 5 forms a LAN together with the computers 1 and 2,
It is a computer for monitoring the operation of the computers 1 and 2. For this monitoring operation, the computer 5 executes a network monitoring manager 5A as an application program.

Computers 1, 2, and 5 are connected to network cable 4 by Ethernet (registered trademark).
et (registered trademark)) standard. Further, TCP / IP is adopted as a communication protocol. Computers 1, 2, and 5 are individually assigned IP addresses (hereinafter, referred to as real IP addresses). Further, an IP address (hereinafter referred to as a virtual IP address) is individually assigned to each of the application programs 1B and 2B executed by the computers 1 and 2. For explanation of the operation, it is assumed that the real IP address of the computer 1 is IP1A, the virtual IP address of the application program 1B is IP1B, the real IP address of the computer 2 is IP2A, and the virtual IP address of the application program 1B is IP2B.

In the network system having such a configuration, one of the computers 1 and 2 in the cluster device 3 is configured to use the application program 1B or 2B.
Execute Which of the computers 1 and 2 executes the application program 1B or 2B is automatically determined by the computer 1 or 2 executing the corresponding cluster program 1A or 2A.

First, assuming that the computer 1 starts executing the application program 1B, the computer 1 issues a trap notification (100) indicating a virtual IP address as shown in FIG. 2 in a startup processing routine of the application program 1B. (Step S
1), and a trap notification indicating the real IP address (101)
Is issued (step S2).

When the computer 1 terminates the execution of the application program 1B, the computer 1 issues a trap notification (102) indicating a virtual IP address in the termination processing routine of the application program 1B as shown in FIG. Then, a trap notification (103) indicating the real IP address is issued (step S4).

These trap notifications (100) to (103) are data indicating the operation state of the computer 1, and (100) to (103) indicate the types of trap notifications. Each time a trap notification is issued, the trap notification is transmitted to the computer 5 via the network cable 4.

When the computer 2 starts executing the application program 2B, it issues a trap notification (100) indicating a virtual IP address as shown in FIG. 2, and further issues a trap notification (101) indicating a real IP address. . Also, when the computer 2 terminates the execution of the application program 2B, it issues a trap notification (102) indicating the virtual IP address as shown in FIG.
Further, a trap notification (103) indicating the real IP address is issued.

The computer 5 accepts each of the above trap notifications by executing the network monitoring manager 5A and determines the contents. That is, the computer 5 receives the trap notifications (100) and (101) as shown in FIG.
Is determined (step S11). When the trap notifications (100) and (101) are received, the virtual IP address indicated by the trap notification (100) is
Is determined (step S12). If the virtual IP address indicated by the trap notification (100) is IP1B, it is determined whether the real IP address indicated by the trap notification (101) is IP1A (step S13).
If it is determined in step S13 that the real IP address is IP1A, it is determined that the computer 1 is executing the application program 1B, and the flag F1 is set to 1 (step S14).

In step S12, a trap notification is issued.
If it is determined that the virtual IP address indicated by (100) is not IP1B, the virtual IP address indicated by the trap notification (100)
It is determined whether the address is IP2B (step S15). If the virtual IP address indicated by the trap notification (100) is IP2B, it is determined whether or not the real IP address indicated by the trap notification (101) is IP2A (step S16). If it is determined in step S16 that the real IP address is IP2A, it is determined that the computer 2 is executing the application program 2B, and the flag F2 is set to 1 (step S16).
17).

In step S11, a trap notification is issued.
If it is determined that it is not the reception of (100) and (101), it is determined whether or not the trap notifications (102) and (103) have been received (step S18). Trap notification (10
When receiving (2) and (103), the trap notification (1
It is determined whether the virtual IP address indicated by 02) is IP1B (step S19). Trap notification (102)
If the virtual IP address indicated by is the IP1B, it is determined whether or not the real IP address indicated by the trap notification (103) is the IP1A (step S20). Step S2
If it is determined at 0 that the real IP address is IP1A, the computer 1 ends the execution of the application program 1B and resets the flag F1 to be equal to 0 (step S21).

In step S19, a trap notification is issued.
If it is determined that the virtual IP address indicated by (102) is not IP1B, the virtual IP address indicated by the trap notification (102)
It is determined whether the address is IP2B (step S22). If the virtual IP address indicated by the trap notification (102) is IP2B, it is determined whether or not the real IP address indicated by the trap notification (103) is IP2A (step S23). If it is determined in step S23 that the real IP address is IP2A, the computer 2 terminates the execution of the application program 2B and resets the flag F2 to be equal to 0 (step S24).

By such an operation, the computer 5 confirms the operation state of the computers 1 and 2.

In the above embodiment, F1 =
Steps S19 and S20 may be executed only when F1 = 1, and steps S23 and S24 may be executed only when F2 = 1.

In the above embodiment, the cluster device 3 is provided with two computers 1 and 2. However, three or more clustered computers may be provided in the cluster device 3. .

In the above embodiment, the trap notifications (100) and (101) are issued individually.
It may be issued as one notification. This is a trap notification
The same applies to (102) and (103).

Furthermore, in the above-described embodiment, TCP / IP is used as a communication protocol. However, the present invention is not limited to this, and another communication protocol may be used.

[0024]

As described above, according to the present invention, it is possible to reliably determine which of a plurality of computers constituting a cluster device is executing an application program.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a network system to which a network monitoring method according to the present invention is applied.

FIG. 2 is a flowchart showing a routine at the start of execution of an application program.

FIG. 3 is a flowchart illustrating a routine at the end of execution of an application program.

FIG. 4 is a flowchart showing the operation of the network monitoring manager.

[Explanation of symbols]

 1,2,5 computer 3 cluster device 4 network cable

Claims (2)

[Claims] 1. A monitoring apparatus comprising a plurality of computers connected to a network, wherein one of the plurality of computers monitors a cluster device that executes an application program predetermined for each of the plurality of computers. A network monitoring method, wherein an address is assigned in advance to each of the plurality of computers and each application program, and each of the plurality of computers assigns an address of the application program and an assigned address of the computer itself when execution of the application program starts. Issue to the monitoring device, the monitoring device obtains the assigned address of the application program and the assigned address of the computer itself in response to the first notification, and updates the application. A network monitoring method characterized by determining a computer that executes an application program. 2. When the execution of the application program is completed, each of the plurality of computers issues a second notification indicating the assigned address of the application program and the assigned address of the computer itself to the monitoring device. 2. The network monitoring method according to claim 1, further comprising: obtaining an assigned address of the application program and an assigned address of the computer itself in response to the second notification, and determining the application program execution end computer.