JPH11175489A - Failure information system by alternative information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To acquire failure information from other communication paths even if a certain communication path is disenabled in communication by allowing a computer which receives the failure information to perform simultaneous broadcast communication to the other computers connected to the self-computer through a communication path thereby giving an alternative information. SOLUTION: Although a computer 1 gives information (simultaneous broadcast communication) of a failure to computers 2 to 4, when a communication 17 that connects the computer 1 and the computer 4 is currently disenabled in communication, the information is not given to the computer 4. The other computers other than the computer 1 receives failure information. The computers other than the computer 1 gives the failure informs (simultaneous broadcast communication) to the other computers by failure informing parts B 15. However, because the failure information is not given, the computer does not send the information to the other computers. Then, in such a case, the computer 2 sends the failure information to the computers 3 and 4, and the computer 3 sends the failure information to the computers 2 and 4. As a result of this, the computers 2 to 4 receive the failure information. Thus, the failure information is alternatively given through the other communicaters 2 and 3.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to an HA (High Ava
ilability) The present invention relates to a failure notification method using a bypass notification in a system.

[0002]

2. Description of the Related Art A system is constructed by connecting a plurality of computers, and when a failure occurs in one computer, tasks to be executed by the computer are distributed to other computers.
An HA system capable of continuously performing an operation is known.

[0003] For example, as shown in FIG.
The system connects the computers and when a failure occurs,
It is known to notify another computer directly connected to a computer in which a failure has occurred.

[0004]

If a communication path becomes uncommunicable for some reason, such as a cable for communication between computers being disconnected, computers at both ends of the communication path cannot know each other's fault. There's a problem.

SUMMARY OF THE INVENTION The present invention has been made in view of the above situation, and provides a highly reliable failure notification system by performing detour notification using three or more computers and a communication path between the computers. Aim.

[0006]

To achieve the above object, according to a first aspect of the present invention, in a network system in which a plurality of computers are connected via a communication path, each of the computers includes its own computer. Fault detecting means for detecting a fault in the computer; fault notifying means for notifying a fault of the own computer to another computer via the communication path; fault receiving means for receiving a fault notification from another computer; Means for notifying another computer that the failure has occurred in any of the plurality of computers by the failure detection means, and the other plurality of computers connected via the communication path by the notification means. The computers that simultaneously broadcast the failure notification to the computer and receive the failure notification by the failure receiving unit communicate with their own computers by the notification unit to the other computers. Characterized in that it is constituted by a fault notification method of performing bypass notification by broadcast simultaneously connected another computer with.

According to a second aspect of the present invention, each of the computers has means for storing a failure notification broadcasted by its own computer, and the failure notification received from another computer is simultaneously broadcasted by its own computer. If the received failure notification is the same as the reported failure notification, the received failure notification is further provided with means for not simultaneously broadcasting the received failure notification to another computer.

According to the present invention, when a failure occurs in the own computer, each computer broadcasts the failure information to other computers at the same time, and the other computer receiving the failure information further transmits the information to its own computer. Broadcast to other computers connected via the communication path. For this reason, each computer is notified of the failure information by a detour notification from another computer in addition to the direct notification of the failure information from the computer in which the failure has occurred. Can receive the failure information via the communication path. For this reason, a highly reliable failure notification system can be provided.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram of an HA (High Availability) system to which the failure notification system of the present invention is applied. The HA system shown in FIG. 1 is constituted by four computers 1, 3, 5, and 7 as an example, but can be constituted by an arbitrary number. Each of the computers 1, 3, 5, and 7 includes a detecting unit 9 for detecting a fault in the own computer, a fault notifying unit A11 for notifying the fault of the own computer to another computer, and a fault receiving unit 13 for receiving a fault of the other computer. And a notifying unit B15 for notifying the received fault information to another computer. Each of the computers 13, 5, and 7 is connected to another computer via the communication path 17.

FIGS. 2 to 5 are a flowchart and a conceptual diagram showing the operation of the failure notification system according to the present invention. Less than,
The operation of the failure notification system according to the present invention will be described with reference to FIGS. Now, in step S1 of FIG.
It is assumed that a failure has occurred in the computer 1. This failure is detected by the failure detection unit 9 of the computer 1 in step S3. In this state, it is assumed that the communication path 17 connecting the computer 1 and the computer 4 is in a communication disabled state as shown in FIG. In step S5 of FIG.
Notifies the directly connected computer of the failure by the failure notification unit A11. That is, as shown in FIG. 4, the computer 1 notifies the computer 2, the computer 3, and the computer 4 of a failure (simultaneous broadcast), but the communication path 17 connecting the computer 1 and the computer 4 cannot communicate anymore. This failure notification is notified only to the computer 2 and the computer 3, but not to the computer 4.
In step S7, the computers other than the computer 1 receive the failure notification from the failure receiving unit 13. However, as described above, only the computers 2 and 3 receive the failure notification, and the computer 4 cannot receive the failure notification. In step S9, the computers other than the computer 1 perform a failure notification (broadcast) to other computers by the failure notification unit B15. However, the computer 4 does not notify other computers because the failure information is not notified. Therefore, in this case, as shown in FIG. 5, the computer 2 notifies the computer 3 and the computer 4 of a failure, and the computer 3 notifies the computer 2 and the computer 4 of a failure. As a result, step S11
, The computers 2, 3, and 4 receive the failure information. As described above, when the communication path is functioning normally, the failure information is directly received from the failed computer, and when the communication path is not available, the failure information issued from the failed computer is received. A detour is notified via another computer. Therefore, if there is at least one normally functioning communication path, failure information can be received by detour notification, and a highly reliable failure notification method can be obtained.

As described above, according to the above-described embodiment, in the HA system in which a plurality of computers are connected, the computer in which the failure has occurred transmits the failure information to all the other computers connected to the own computer via the communication path. (Broadcast).
Each computer that has received the failure information further broadcasts simultaneously to other computers connected to its own computer. The above simultaneous broadcast is repeated until the failure information is notified to all the computers except the computer in which the failure has occurred. In this case, it is conceivable that the notification of the broadcast is looped between the computers, but this can be prevented by, for example, having a transmission record table in each node.

FIGS. 6 to 9 are conceptual diagrams showing another embodiment of the present invention. In the first embodiment shown in FIGS.
The A system is composed of four units. As mentioned above,
In the case of notifying the detour of the fault information, it is desirable to notify all the computers with the minimum number of steps. The second embodiment shown in FIGS. 6 to 9 shows an HA system constituted by eight computers. If there is a restriction on the number of communication paths that can be connected to one computer, it is possible to notify all computers with a minimum number of steps by optimizing the connection form. For example, when three communication paths can be connected to one computer, four computers can be connected one-to-one with other computers, but in an eight-computer configuration, they can be connected one-to-one. Cannot connect. The second embodiment shows an example in which fault information can be notified with a minimum detour.

As shown in FIG. 6, each computer is connected to a star (star network). Each computer is connected to an adjacent computer and an opposite computer. For example, the computer 21 is connected to the computers 23 and 35 and also to the computer 29. In such a configuration, when each communication line is functioning normally, the failure information can be notified up to twice, and even when the communication line is disabled, all the computers can communicate with each other up to three times. Can be notified of failure information. Hereinafter, a specific example will be described.

Now, assuming that a failure has occurred in the computer 35 and all communication paths are normal, the computer 35
1, 33 and 27 are notified of the fault information directly. At this stage, the failure information has not been notified to the computers 23, 25, 29, 31. Next, the computer 21 is connected to the computers 23, 2
9, the computer 33 notifies the computers 31, 25,
The computer 27 notifies the computers 25 and 29. As a result,
The computers 21, 33, and 27 are notified of the failure by one inter-computer communication, and the computers 23, 25, 29, and 31 are notified of the failure by two inter-computer communications. Therefore, when each communication path is operating normally, a failure can be notified by a maximum of two inter-computer communications.

On the other hand, as shown in FIG. 6, it is assumed that the communication path 17 is now disabled. In this case, the computer 35 notifies the computers 21 and 33 of the failure as shown in FIG. In the second inter-computer communication, the computer 21 notifies the computer 23 and the computer 29 of the failure information, and the computer 33 sends the computer 31
And notifies the computer 25 of the failure information. Therefore,
Only the computer 27 is not notified of the failure in the second inter-computer communication. In the third inter-computer communication,
As shown in FIG. 9, the computer 23 includes computers 21, 25, 3
1 is notified of the failure information, and the computer 25
7 and 33, and the computer 29 sends the information to the computers 27, 21, 3
1 and the computer 31 notifies the computers 23, 33 and 29. In FIG. 9, the numbers shown in each computer are:
Shows the number of inter-computer communications until failure information is received. As can be seen from the figure, according to the present invention, even if a communication path becomes unavailable, fault information can be notified to all computers by a maximum of three inter-computer communications by a maximum of three inter-computer communications. .

[0016]

As described above, according to the present invention, in addition to the direct notification of the fault information from the failed computer, a detour notification from another computer can be received. Even when communication becomes unavailable, a failure notification can be obtained from another communication path, and a highly reliable failure notification can be made.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an HA system to which a failure notification method according to the present invention is applied.

FIG. 2 is a flowchart showing an operation of a failure notification in the HA system shown in FIG. 1;

FIG. 3 is a conceptual diagram showing a failure notification operation when a failure occurs in a computer and a communication path becomes unavailable in the HA system shown in FIG. 1;

FIG. 4 is a conceptual diagram showing a failure notification operation when a failure occurs in a computer and a communication path becomes unavailable in the HA system shown in FIG. 1;

FIG. 5 is a conceptual diagram showing a failure notification operation when a failure occurs in a computer and a communication path becomes unavailable in the HA system shown in FIG. 1;

FIG. 6 is a conceptual diagram showing another embodiment of the present invention;
FIG. 4 is a conceptual diagram showing a failure notification operation when a failure occurs in an HA system including two computers.

FIG. 7 is a conceptual diagram showing another embodiment of the present invention;
FIG. 4 is a conceptual diagram showing a failure notification operation when a failure occurs in an HA system including two computers.

FIG. 8 is a conceptual diagram showing another embodiment of the present invention.
FIG. 4 is a conceptual diagram showing a failure notification operation when a failure occurs in an HA system including two computers.

FIG. 9 is a conceptual diagram showing another embodiment of the present invention;
FIG. 4 is a conceptual diagram showing a failure notification operation when a failure occurs in an HA system including two computers.

FIG. 10 is a conceptual diagram for explaining a failure notification method in a conventional HA system.

[Explanation of symbols]

1, 3, 5, 7 Computer 9 Fault detection unit 11 Fault notification unit 13 Fault reception unit 15 Fault notification unit B 17 Communication path 21, 23, 25, 27, 29, 31, 33, 35 …calculator

Claims (2)

[Claims] In a network system in which a plurality of computers are connected via a communication path, each of the computers detects a fault in the own computer; and a fault detecting means for detecting a fault in the own computer via the communication path. A failure notification unit for notifying the other computer; a failure receiving unit for receiving a failure notification from another computer; and a unit for notifying the received failure notification to another computer, and a failure has occurred in any of the plurality of computers. The failure detecting means detects the failure, broadcasts a failure notification to the other computers connected via the communication path by the notification means, and receives the failure notification by the failure receiving means. A failure notification method that performs a detour notification by simultaneously broadcasting to another computer connected to the own computer via a communication path by a notification means to the other computer. formula. 2. Each of the computers has means for storing a failure notification broadcasted by its own computer, and a failure notification received from another computer is the same as the failure notification broadcasted by its own computer. 2. The failure notification system according to claim 1, further comprising means for not simultaneously broadcasting the received failure notification to another computer.