JP2004355446A - Cluster system and its control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the reliability of a cluster system by arranging one auxiliary server for a plurality of working servers, and when a fault is generated in one of the working servers, switching the failed working server to the the auxiliary server by the lead of the auxiliary server. <P>SOLUTION: Each server is provided with an operation setting part, an operation display part, a fault communication part, and a system control part. A plurality of working servers and one or more auxiliary servers are arranged to constitute a cluster system. The working server monitors its own server's fault, and when a fault is generated, reports fault generation to the fault communication part. The auxiliary server monitors the fault communication parts of the working servers, and when the generation of a fault in one of the working servers is confirmed, shuts down the failed working server by the lead of the auxiliary server and switches the failed working server to the auxiliary server itself. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a cluster system configured with a plurality of servers and having increased reliability and a control method thereof, and more particularly to a cluster that switches from an active server to a standby (standby) server when a failure occurs in an active (active) server. The present invention relates to a system and a control method thereof.
[0002]
[Prior art]
With the spread of server systems in recent years, higher reliability is required for server systems. Normally, to increase the reliability of a server system, a redundant configuration that prepares a spare server that is not normally operating is often used so that even if a failure occurs in the active server, the entire system can be operated without going down. Has been adopted.
[0003]
A commonly used conventional server system having a redundant configuration is a system in which a server has a redundant configuration. In this case, an active server and a spare server to be paired with the active server are prepared, and the states are mutually monitored. When a failure occurs in the active server, the spare server takes over for the active server and takes over the work.
[0004]
As another switching method at the time of failure, as in the method of Patent Document 1, an active server, a spare server, and a management server are arranged, the management server monitors the state of the active server, and a failure occurs in the active server. There is a method of switching to a spare server at the initiative of the management server.
[Patent Document 1]
JP 2000-347959 A
[Problems to be solved by the invention]
However, in the conventional dual configuration method, a spare server is required for each active server, so twice as many resources are required and the cost is increased. Further, it is not possible to determine which server is in which operation state by appearance, and maintenance becomes difficult when the number of servers increases.
[0006]
Further, the method described in Patent Document 1 requires a management server in addition to the spare server, and is inefficient because the spare server idles until a failure occurs in the active server.
[0007]
The present invention has been made in view of the above-described problems, and has as its object to provide a cluster system including a plurality of servers, having high reliability, and suppressing an increase in cost, and a control method thereof.
[0008]
Another object of the present invention is to provide a cluster system which is composed of a plurality of servers, has high reliability, can easily determine the operation state of each server from the external appearance, and can be easily maintained, and a control method therefor.
[0009]
Another object of the present invention is to provide a highly reliable and efficient cluster system including a plurality of servers and a control method thereof.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, in the cluster system according to the present invention, one or more spare servers are arranged for a plurality of active servers providing the same service, and when a failure occurs in the active server, the spare server is provided. The server is switched from the active server to the spare server at the initiative of the server. In other words, a plurality of servers having the same configuration and function and capable of switching between the active and standby operating states are connected to a network, and when a failure occurs in the active server that is operating as the active server, the active server is stopped and the standby server is stopped. A standby system that is operated as a standby server and switches as the active server, wherein the plurality of servers are set as the active server and a smaller number of standby servers than the active server, and the standby server is initiative when the active server fails. The backup / active server is switched.
[0011]
The cluster system according to the present invention is a cluster system configured by connecting a plurality of servers to a network, wherein each of the servers includes an operation setting unit that sets an active / standby operation state and a current operation state. An external operation display unit, and an active monitoring unit that functions as a standby server when the operating status is set to “spare” and periodically monitors the presence or absence of a failure of the active server whose operating status is set to “active” A failure detection unit that monitors the failure of the server when the operation state is set to “working” and indicates whether there is a failure, and a system control unit that performs shutdown control and reboot control of the server. A plurality of active servers are monitored for failures by a smaller number of spare servers than the active servers. When a failure of the active server is detected, the active servers are shut down and one of the spare servers is shut down. And switches as the active server.
[0012]
In order to achieve the above and other objects, in the cluster system of the present invention, when the spare server is switched to the active server in the event of a failure of the active server, the fact that the spare server has been switched to the active server and the fact that the active server has been shut down. Provide a display that can be easily identified from the appearance.
[0013]
That is, in the above-described configuration, when the operation display unit of each server is operating as the active server, the display indicating the active operation is performed, and when operating as the standby server, the display indicating the standby operation is displayed. When stopped, a display indicating a failure stop is displayed on the operation display section.
[0014]
As a result, even after the server operation is switched due to the occurrence of a failure, it is possible to easily determine from the appearance which server is currently in use, standby, or stopped due to a failure, thereby facilitating maintenance.
[0015]
Further, in the cluster system according to the present invention, a smaller number of spare servers are allocated to a plurality of active servers, and when the active server fails, the servers can be switched under the initiative of the spare server. This makes it possible to perform switching without having to prepare a management server in addition to the spare server.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. First, a cluster system according to a first embodiment of the present invention will be described with reference to FIGS.
[0017]
As shown in FIG. 1, the cluster system according to the first embodiment of the present invention includes a first active server 100, a second active server 200, and a spare server that have the same components and can provide the same service. 300 is connected to a network by a LAN 400 and a control bus 500.
[0018]
Note that the components of the second active server 200 are the same as those of the first active server, so that the details are omitted in FIG.
[0019]
Here, the active server in the present invention is a server that performs a task of providing a service to a client, and a spare server is a server that does not perform a task of providing a service to a client. According to the present invention, a plurality of servers constituting the cluster system are set as an active server and a smaller number of spare servers than the active server.
[0020]
The first active server 100 (and the second active server 200) has an operation setting unit 120 for setting in advance that an active operation is to be performed at the time of start-up, and a display for easily determining from the appearance that the active operation is being performed. The system includes an operation display unit 110 that performs the operation, a system control unit 140 that controls the shutdown and reboot of the system, and a failure communication unit 130 that detects a failure and notifies a failure.
[0021]
Similarly, the spare server 300 also includes an operation setting unit 320 for setting in advance that a preliminary operation is to be performed at the time of startup, an operation display unit 310 for displaying a display that allows the user to easily determine from the appearance that the preliminary operation is being performed, and a system. A system control unit 340 for controlling shutdown and reboot of the system, and a failure communication unit 330 for detecting a failure and notifying a failure.
[0022]
The failure communication units 130 and 330 include active monitoring units 131 and 331 and failure detection units 132 and 332, and change their movements according to the contents set in the operation setting units 120 and 320. When the operation setting is “working”, the failure detection unit is activated. The failure detection unit monitors a failure of its own server (active server) and, when a heartbeat is monitored from the spare server, notifies failures such as a disk error, a memory error, and a power supply error of the own server and a heartbeat response. When the operation setting is “spare”, the active monitoring unit is activated. The active monitoring unit periodically accesses the failure detection unit of the active server and monitors whether the active server has a failure.
[0023]
Next, the operation of the first embodiment will be described with reference to FIG. 2 illustrating the initial setting operation, FIG. 3 illustrating the same flowchart, FIG. 4 illustrating the switching operation in the event of a failure of the active server, and FIG. . For simplicity, the configuration will be described with two active servers and one spare server.
[0024]
First, as an initial setting, as shown in FIGS. 2 and 3, an operation initial setting is input to the first active server 100 (and the second active server 200) (P10 in FIG. 3). As a setting method, for example, setting by a dip switch, setting by software command input, and the like may be used. The operation setting unit 120 stores the input information (P20 in FIG. 3), and the operation display unit 110 displays the operation based on the settings saved by the operation setting unit 120 (P30 in FIG. 3).
[0025]
FIG. 10 shows a specific example of the operation display unit 110. For example, the operation states are defined as five states of “working operation”, “preliminary operation”, “preliminary 2 operation”, “failure stop”, and “stop”, and are indicated by five LEDs.
[0026]
The failure communication unit 130 makes an operation determination based on the settings saved by the operation setting unit 120 (P40 in FIG. 3). In the case of the active operation, the failure of the local server (active server) is monitored, and when a heartbeat is monitored from the spare server, a failure notification and a heartbeat response such as a disk error, a memory error, and a power error of the local server are performed. The failure detection unit 132 to be activated is activated (P50 in FIG. 3).
[0027]
Similarly, the spare server 300 performs the operations from P10 to P40 in FIG. 3, and in the case of the spare operation, the active monitoring unit 331 that periodically accesses the failure detection unit of the active server and monitors the active server for a failure is provided. Activate (P60 in FIG. 3).
[0028]
Next, a switching method when a failure occurs in the active server will be described. As shown in FIGS. 4 and 5, the failure detection unit 132 of the first active server 100 (and the second active server 200) monitors the occurrence of a failure such as a disk error, a memory error, and a power supply error of its own server. A failure is detected (S10 in FIG. 5).
[0029]
The active monitoring unit 331 of the spare server 300 accesses the failure detecting unit 132 of the first active server 100 (and the second active server 200) at a fixed cycle, and accesses the first active server 100 (and the second active server 200). ), And performs heartbeat monitoring (S20 in FIG. 5).
[0030]
When the active monitoring unit 331 of the spare server 300 detects the failure of the first active server 100 (Yes route in S30 of FIG. 5), the first active server 100 is shut down, and the spare server 300 itself becomes the active server. In order to switch, the active monitoring unit 331 of the spare server 300 issues a shutdown instruction to the failure detecting unit 132 of the first active server 100 (S40 in FIG. 5).
[0031]
Upon receiving the shutdown instruction from the active monitoring unit 331 of the spare server 300, the failure detection unit 132 of the first active server 100 enables the operation display unit 110 to easily determine from the appearance that the operation state has been switched. The display setting is changed from “active” to “failure stop” (S50 in FIG. 5), a shutdown instruction is issued to the system control unit 140 of the first active server 100, and the system control unit 140 of the first active server 100 is shut down. Executes the shutdown of the first active server 100 (S60 in FIG. 5).
[0032]
After issuing a shutdown instruction to the failure detection unit 132 of the first active server 100, the standby server 300 changes the display setting of the operation setting unit 320 of the standby server 300 from “spare” to “active” (FIG. 5). S70). Then, the operation display unit 310 of the spare server 300 receives the setting change of the operation setting unit 320, changes the operation display from “spare” to “active” (S80 in FIG. 3), and starts operation as the active server.
[0033]
In the first embodiment, the case of two active servers and one spare server has been described as an example. However, the present invention can be implemented with N active servers.
[0034]
According to the procedure of the first embodiment described above, in a cluster system composed of a plurality of servers and having increased reliability, fewer spare servers are allocated to a plurality of active servers, and the spare server The server can be switched under the initiative, and the switching can be performed without preparing a management server in addition to the spare server. This makes it possible to provide a cluster system with high reliability and reduced cost. Further, even if there are a plurality of servers and the server operation is switched due to the occurrence of a failure, it is possible to easily determine which server is currently in use, a backup, or a failure, from the appearance, thereby facilitating maintenance.
[0035]
Next, a cluster system according to a second embodiment of the present invention will be described with reference to FIGS.
[0036]
In a second embodiment, in a cluster system composed of a plurality of servers and having increased reliability, an operation setting unit for setting an active / standby operation state in the server and an operation display unit for externally displaying the current operation state. When the operation state is set to “spare”, an active monitoring unit that periodically monitors the presence or absence of a failure of the active server whose operation state is set to “active”, and the operation state is set to “active”. In this case, the system includes a failure detection unit that monitors the failure of the own server (active server) and indicates whether there is a failure, and a system control unit that performs shutdown control and reboot control of the own server. Monitors the presence or absence of a failure, reboots the active server when the first failure of the active server that has been monitored for failure is detected, and shuts down the same active server when the same active server failure is detected again after the reboot. A cluster system and switches the standby server down as the active server.
[0037]
In the second embodiment, when a failure occurs in an active server that is operating in a cluster system having a plurality of servers and having increased reliability, the active server is stopped and a standby server that is on standby is operated. In the method of switching as an active server, a step of executing a failure monitoring of an active active server during a standby (standby) operation, and a step of instructing a reboot of the active server when a first failure of the active server is detected. The step of instructing shutdown of the active server when the failure of the active server is detected again after the reboot, the step of changing the operation setting of the spare server from “spare” to “active”, and displaying the operation display of the spare server. A cluster system comprising: a step of changing based on an operation setting; and a step of operating by switching from a preliminary operation to a working operation. A server switching method systems out.
[0038]
The basic configuration and the initial setting operation of the cluster system according to the second embodiment are the same as those of the first embodiment. Regarding the difference of the second embodiment from the first embodiment, FIG. 6 is an explanatory view of the switching operation when the active server fails, and FIG. For simplicity, the configuration will be described with two active servers and one spare server.
[0039]
The failure detection unit 132 of the first active server 100 (and the second active server 200) monitors the occurrence of a failure such as a disk error, a memory error, or a power supply error of the own server (active server) and detects the occurrence of the failure ( S10 in FIG. 7).
[0040]
The active monitoring unit 331 of the spare server 300 accesses the failure detecting unit 132 of the first active server 100 (and the second active server 200) at a fixed cycle, and accesses the first active server 100 (and the second active server 200). ) Is performed and the heartbeat monitoring is performed (S20 in FIG. 7).
[0041]
When the active monitoring unit 331 of the spare server 300 detects a failure of the first active server 100 (Yes route of S30 in FIG. 7), the active monitoring unit 331 of the spare server 300 performs the operation on the first active server 100. The number of failure detections is confirmed (S31 in FIG. 7). When the first active server 100 detects the first failure, the active monitor 331 of the spare server 300 sends the failure to the failure detection unit 132 of the first active server 100 in order to reboot the first active server 100. In response, a reboot instruction is issued (S32 in FIG. 7).
[0042]
Upon receiving the reboot instruction from the active monitoring unit 331 of the spare server 300, the failure detection unit 132 of the first active server 100 instructs the system control unit 140 of the first active server 100 to perform a reboot, and the system control unit 140 Executes the reboot of the first active server 100 (S33 in FIG. 7).
[0043]
When the failure detection unit 132 of the first active server 100 detects a failure again after the first active server 100 is rebooted (S10 ′ to S30 in FIG. 7), the active monitoring unit 331 of the spare server 300 sets The number of times of failure detection in the first active server 100 is confirmed (S31 in FIG. 7). When the failure detection number of the first active server 100 is the second time, the first active server 100 is shut down and the spare server 300 itself is switched to the active server. A shutdown instruction is issued to the failure detection unit 132 of the active server 100 (S40 in FIG. 7).
[0044]
Upon receiving the shutdown instruction from the active monitoring unit 331 of the spare server 300, the failure detection unit 132 of the first active server 100 enables the operation display unit 110 to easily determine from the appearance that the operation state has been switched. The display setting is changed from “active” to “failure stop” (S50 in FIG. 7), a shutdown instruction is issued to the system control unit 140 of the first active server 100, and the system control unit 140 of the first active server 100 is shut down. Executes the shutdown of the first active server 100 (S60 in FIG. 5).
[0045]
After issuing a shutdown instruction to the failure detection unit 132 of the first active server 100, the spare server 300 changes the display setting of the operation setting unit 320 of the standby server 300 from "spare" to "active" (FIG. 7). S70). Then, the operation display unit 310 of the spare server 300 receives the setting change of the operation setting unit 320, changes the operation display from “spare” to “active” (S80 in FIG. 7), and starts operation as the active server.
[0046]
In the second embodiment, the case of using two active servers and one spare server has been described as an example, but it is naturally possible to use N active servers.
[0047]
According to the procedure of the second embodiment described above, the failure of the active server that can be recovered by rebooting, which is often seen in practical use, is dealt with only by rebooting the active server. For a required failure, the spare server can be switched to the active server.
[0048]
Next, a specific example in which a plurality of spare servers in the first and second embodiments are added will be described as a third embodiment of the present invention with reference to FIGS. explain. When there are a plurality of spare servers, for example, a state of “spare 2” meaning “spare spare” is newly defined as an operation state.
[0049]
FIG. 8 is an explanatory view of switching the operation of a plurality of spare servers in the event of a failure of the active server, and FIG. For simplicity, one active server and two spare servers are used, and additional contents from the first embodiment and the second embodiment will be described. The basic configuration and the initial setting operation of the third embodiment are the same as those of the first embodiment and the second embodiment. Only the initial setting of the second spare server 600 is added.
[0050]
Regarding the operation of the third embodiment, S10 to S80 of FIGS. 4 and 5 of the first embodiment and S10 to S80 of FIGS. 6 and 7 of the second embodiment are the same as those of the third embodiment. It is. After that, in the third embodiment, since there are two spare servers (the first spare server 300 and the second spare server 600), additional processing is performed.
[0051]
That is, in S80 of FIG. 9, similarly to S80 of the first or second embodiment, the operation display unit 310 of the first spare server 300 receives the setting change of the operation setting unit 320 and displays the operation display. From “spare” to “working”.
[0052]
Subsequent to this processing, the active monitoring unit 331 of the first standby server 300 issues an instruction to the active monitoring unit 631 of the second standby server 600 to switch from the “standby 2” operation to the “standby” operation (FIG. 9). S90). When receiving the instruction to switch to the “spare” operation, the active monitoring unit 631 of the second spare server 600 changes the display setting of the operation setting unit 620 of the second spare server 600 from “spare 2” to “spare”. (S100 in FIG. 9). Then, the operation display unit 610 of the second spare server 600 receives the setting change of the operation setting unit 620, changes the operation display from “spare 2” to “spare” (S110 in FIG. 9), and starts operation as a spare server. I do.
[0053]
In this example, the case of one active server and two spare servers has been described as an example. However, a plurality of active servers and a plurality of spare servers (smaller than the active server) can be easily expanded. It is possible.
[0054]
According to the third embodiment, by preparing a plurality of spare servers, even if a failure occurs in a plurality of active servers at one time, the spare server can be switched to the active server, and the system suspension period can be shortened. It is.
[0055]
FIG. 10 is a specific example of the operation display unit 110 in each embodiment of the present invention, and the operation states are “active operation”, “preliminary operation”, “secondary operation”, “failure stop”, and “stop”. "5 states are indicated by five LEDs. According to this embodiment, even when the active server and the spare server are exchanged, it is possible to easily determine which operating state is in the external appearance, so that maintenance is easy even when the number of servers is large.
[0056]
FIG. 11 is another specific example of the operation display unit 110 in each embodiment of the present invention. For example, the operation states are defined as five states of “active operation”, “preliminary operation”, “preliminary 2 operation”, “failure stop”, and “stop”, and are indicated by three LEDs. According to this embodiment, even when the active server and the spare server are exchanged, it is possible to easily determine which operating state is in the external appearance, so that maintenance is easy even when the number of servers is large. Further, the number of LEDs of the operation display unit can be reduced.
[0057]
In FIG. 10 and FIG. 11, as a specific example of the operation display unit 110 in each embodiment of the present invention, the operation state is determined by giving meaning to turning on and off of the LED, but the blinking cycle of the LED is similarly combined. Naturally, it can be determined by the display method. Also, it is naturally possible to mount a liquid crystal screen instead of the LED and display the operation state by character information.
[0058]
【The invention's effect】
As described above, according to the present invention, in a cluster system in which one or more spare servers are allocated to a plurality of active servers, the spare server has the function of a management server, and the spare server has the function of the management server. It is possible to switch from the active server to the spare server without requiring a server. It has a simple configuration that does not require a management server, and can suppress an increase in system cost.
[0059]
Each server according to the present invention includes an operation setting unit that sets in advance whether to operate as a current or a standby at the time of startup, an operation display unit that allows the current operation mode to be easily determined from the appearance, a system shutdown and a reboot. And a failure communication unit that monitors the presence or absence of a failure and notifies the failure. Then, by the cooperation of these servers, switching from the active server to the spare server can be performed under the initiative of the spare server. When a failure occurs in the active server, switching can be promptly performed under the initiative of the spare server, and reliability can be improved.
[0060]
Further, the current operation mode can be easily determined from the appearance by the operation display unit provided in each server, and which server is in which operation state easily from the appearance even after switching the operation of the active server and the spare server. Can be determined. Therefore, maintenance is easy.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a cluster system according to a first embodiment (and second and third embodiments) of the present invention.
FIG. 2 is an initial operation diagram according to the first embodiment (and the second and third embodiments) of the present invention.
FIG. 3 is an initial operation flowchart according to the first embodiment (and the second and third embodiments) of the present invention.
FIG. 4 is a failure switching operation diagram according to the first embodiment of the present invention.
FIG. 5 is a failure switching flowchart according to the first embodiment of the present invention.
FIG. 6 is a failure switching operation diagram according to the second embodiment of the present invention.
FIG. 7 is a failure switching flowchart according to the second embodiment of the present invention.
FIG. 8 is a failure switching operation diagram according to the third embodiment of the present invention.
FIG. 9 is a failure switching flowchart according to the third embodiment of the present invention.
FIG. 10 is a specific example of an operation display unit in each embodiment of the present invention.
FIG. 11 is another specific example of the operation display unit in each embodiment of the present invention.
[Explanation of symbols]
100 first active server 110 operation display unit 120 operation setting unit 130 failure communication unit 131 active monitoring unit 132 failure detection unit 140 system control unit 200 second active server 300 spare server (first spare server)
310 operation display unit 320 operation setting unit 330 failure communication unit 331 active monitoring unit 332 failure detection unit 340 system control unit 400 LAN
500 control bus 600 second spare server 610 operation display unit 620 operation setting unit 630 failure communication unit 631 active monitoring unit 632 failure detection unit.

Claims (7)

A plurality of servers having the same configuration and function and capable of switching between active and standby operating states are connected to a network. When a failure occurs in the active active server, the active server is stopped and the standby is activated. A cluster system in which a spare server in operation is operated and switched as an active server,
A cluster system, wherein the plurality of servers are set as an active server and a smaller number of standby servers than the active server, and when the active server fails, the standby server takes the initiative to switch the standby / active server. What is claimed is: 1. A cluster system comprising a plurality of servers connected to a network, wherein each of the servers has an operation setting unit for setting an active / standby operation state, an operation display unit for externally displaying a current operation state, When the status is set to “spare”, the active monitoring unit functions as a spare server and periodically monitors the presence or absence of a failure of the active server whose operation status is set to “current”, and the operation status is set to “active”. When set, a failure detection unit that monitors the failure of the own server and indicates whether there is a failure, and a system control unit that performs shutdown control and reboot control of the own server,
It is characterized in that the presence / absence of a plurality of active servers is monitored by a smaller number of spare servers than the active server, and when a failure of the active server is detected, the active server is shut down and one of the spare servers is switched as the active server. And a cluster system. What is claimed is: 1. A cluster system comprising a plurality of servers connected to a network, wherein each of the servers has an operation setting unit for setting an active / standby operation state, an operation display unit for externally displaying a current operation state, When the status is set to “spare”, the active monitoring unit functions as a spare server and periodically monitors the presence or absence of a failure of the active server whose operation status is set to “current”, and the operation status is set to “active”. When set, the system includes a failure detection unit that functions as an active server, monitors a failure of the own server, and indicates whether there is a failure, and a system control unit that performs shutdown control and reboot control of the own server,
The presence or absence of a failure of the active server is monitored by a smaller number of spare servers than the active server. When the active server detects a failure of its own server during the active operation, the occurrence of the failure is reported to the spare server, and A cluster system which shuts down its own server upon receiving a shutdown instruction or reboots its own server upon receiving a reboot instruction from the spare server. 4. The cluster system according to claim 1, further comprising a display unit having a plurality of display elements in an operation display unit for displaying an operation state of the server, wherein a lighting condition of each display element means an operation state, A cluster system characterized in that the server operation state can be determined from the appearance even after the server operation is switched. A plurality of servers having the same configuration and function and capable of switching between active and standby operating states are connected to a network. When a failure occurs in the active active server, the active server is stopped and the standby is activated. A method for controlling a cluster system in which a middle spare server is operated and switched as an active server,
Setting the plurality of servers as an active server and a smaller number of spare servers than the active server;
A method for controlling a cluster system, comprising the step of switching between a standby server and an active server under the initiative of a standby server when an active server fails. In a cluster system configured by connecting a plurality of servers capable of switching between active / standby operation states to a network, when a failure occurs in an active server that is operating as an active server, the active server stops operating and stands by as a standby server A method for controlling a cluster system in which a middle spare server is operated and switched as an active server,
Setting the plurality of servers as an active server and a smaller number of spare servers than the active server;
Performing failure monitoring of the active working server when the spare server is operating;
Instructing shutdown of the active server when detecting a failure of the active server;
Changing the operation setting of the spare server operating as a spare from “spare” to “working”;
Changing the operation display of the spare server to the outside based on the operation setting;
A method for controlling a cluster system, comprising a step of operating by switching from a standby operation to an active operation. A plurality of servers capable of switching between active / standby operation states are connected to a network, and each of the servers is configured to set an active / standby operation state, and an operation display externally showing the current operation state. , A working monitoring unit, a failure detection unit, and a control method of a cluster system including a system control unit,
Setting the plurality of servers as an active server and a smaller number of spare servers than the active server;
Detecting a failure of the own server during the active operation and indicating a failure to the standby spare server;
Changing the operation display of the running own server from “working” to “failure stop” when the shutdown instruction is received from the spare server and executing the shutdown of the own server;
A method for controlling a cluster system, comprising a step of executing a reboot of a server when a reboot instruction is received from a spare server.

Images