JPWO2019171704A1 - Management server, cluster system, cluster system control method, and program - Google Patents

Abstract

Realize an HA cluster system that can avoid the occurrence of problems caused by the active server and standby server having clusterware. The management server (1) provides the service with a monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server for failover. The transmission processing unit (2) that performs the process of transmitting to the server and the service providing server are requested to execute the monitoring script and return the execution result, and the operating state of the service providing server is determined based on the execution result. A server monitoring processing unit (3) to be monitored, and a cluster control processing unit (4) that requests the service providing server to execute the control script when the execution result of the monitoring script indicates an abnormality of the service providing server. Have.

Description

The present invention relates to a management server, a cluster system, a control method of a cluster system, and a program.

Various technologies have been proposed as technologies related to server monitoring. For example, Patent Document 1 describes a monitored server that is a computer to be monitored, a job monitoring server that is a computer that monitors jobs, and a remote monitoring server that is connected to the job monitoring server via a communication network. The job monitoring system provided is disclosed.

By the way, the process of monitoring the server is also required in the HA (High Availability) cluster system. Generally, in an HA cluster system, the active server and the standby server each have clusterware. Then, the active server and the standby server communicate with each other via the network to form an HA cluster system. In an HA cluster system with such a configuration, the active server monitors whether the local server can normally provide business services, and the standby server monitors whether the local server can take over the business services normally.

In an HA cluster system with such a configuration, in order to monitor business services, application software, middleware, hardware, etc., a monitoring agent process is resident on the active server and the standby server. In addition, it is necessary to put an agent in these servers to deal with an OS (Operating System) panic and to realize functions such as data mirroring. That is, it is necessary to install the clusterware on the active server and the standby server.

Japanese Unexamined Patent Publication No. 2011-159011

As described above, when constructing the HA cluster system described above, it is necessary to install the clusterware on the managed server (active server and standby server). Therefore, when changing an existing system that is not a cluster system to a cluster system, the server is restarted and the system is stopped due to the installation of clusterware and the construction of a cluster. Further, in the above-mentioned HA cluster system, the load due to the processing by the clusterware may become excessive.
On the other hand, the job monitoring system of Patent Document 1 is not an HA cluster system. Therefore, there is still a need for a technique for realizing an HA cluster system that can avoid the occurrence of problems caused by the active server and the standby server having clusterware.

Therefore, one of the objectives to be achieved by the embodiment disclosed in the present specification is an HA cluster system capable of avoiding the occurrence of problems caused by the active server and the standby server having clusterware. It is to provide a management server, a cluster system, a control method of the cluster system, and a program capable of realizing the above.

The management server according to the first aspect includes a monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server for failover. The transmission processing means that performs the process of transmitting to the service providing server and the service providing server are requested to execute the monitoring script and return the execution result, and the operating state of the service providing server is determined based on the execution result. It has a server monitoring processing means to be monitored, and a cluster control processing means that requests the service providing server to execute the control script when the execution result of the monitoring script indicates an abnormality of the service providing server.

The cluster system according to the second aspect performs failover between the active server for providing the predetermined service, the standby server for providing the predetermined service, and the active server and the standby server. The management server includes a management server to be controlled, and the management server has a first monitoring script for monitoring the operation of the active server and a first monitoring script for controlling the operation of the active server regarding failover. A process of transmitting the control script to the active server, a second monitoring script for monitoring the operation of the standby server, and a second monitoring script for controlling the operation of the standby server regarding failover. The transmission processing means that performs the process of transmitting the control script to the standby system server, the active system server, and the standby system server are requested to execute the monitoring script and return the execution result to the execution result. When the server monitoring processing means for monitoring the operating state of the active server and the standby server and the execution result of the monitoring script of the active server indicate an abnormality of the active server, the first It has a cluster control processing means that requests the active server to execute the control script and the standby server to execute the second control script.

In the cluster system control method according to the third aspect, a monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server for failover. Is transmitted to the service providing server, the service providing server is requested to execute the monitoring script and return the execution result, and the operating state of the service providing server is monitored based on the execution result. When the execution result of the monitoring script indicates an abnormality of the service providing server, the service providing server is requested to execute the control script.

The program according to the fourth aspect includes a monitoring script for monitoring the operation of the service providing server that provides a predetermined service, and a control script for controlling the operation of the service providing server for failover. A transmission processing step for performing a process of transmitting to the service providing server, requesting the service providing server to execute the monitoring script and returning the execution result, and monitoring the operating state of the service providing server based on the execution result. When the server monitoring processing step to be performed and the execution result of the monitoring script indicate an abnormality of the service providing server, the computer is made to execute the cluster control processing step that requests the service providing server to execute the control script.

According to the above-described aspect, the management server, cluster system, and cluster system that can realize an HA cluster system that can avoid the occurrence of problems caused by the active server and the standby server having clusterware. Control methods and programs can be provided.

It is a block diagram which shows an example of the structure of the management server which concerns on the outline of embodiment. It is a block diagram which shows an example of the HA cluster system which concerns on Embodiment 1. FIG. It is a block diagram which shows an example of the hardware configuration of the operation management server which concerns on embodiment. It is a flowchart which shows the operation flow of the cluster construction process in the HA cluster system which concerns on Embodiment 1. FIG. It is a flowchart which shows the flow of operation of starting a cluster in the HA cluster system which concerns on Embodiment 1. It is a flowchart which shows the flow of the monitoring operation in the HA cluster system which concerns on Embodiment 1. FIG. It is a flowchart which shows the failover operation in the HA cluster system which concerns on Embodiment 1. FIG. It is a flowchart which shows the operation about execution of the script in the HA cluster system which concerns on Embodiment 1. FIG. It is a block diagram which shows an example of the HA cluster system which concerns on Embodiment 2. FIG. It is a flowchart which shows the operation of the server monitoring and server control using BMC in the HA cluster system which concerns on Embodiment 2. FIG. It is a block diagram which shows an example of the HA cluster system which concerns on Embodiment 3.

<Outline of Embodiment>
Prior to the description of the embodiment, the outline of the embodiment according to the present invention will be described.
FIG. 1 is a block diagram showing an example of the configuration of the management server 1 according to the outline of the embodiment. The management server 1 has a transmission processing unit 2, a server monitoring processing unit 3, and a cluster control processing unit 4. The management server 1 is connected to a service providing server (not shown in FIG. 1) constituting the HA cluster system so as to be capable of wired or wireless communication. The service providing server is, for example, a server that provides a predetermined service to a client device (not shown).

The transmission processing unit 2 performs a process of transmitting a monitoring script for monitoring the operation of the service providing server and a control script for controlling the operation of the service providing server to the service providing server.

The server monitoring processing unit 3 requests the service providing server to execute the monitoring script and return the execution result. Therefore, the service providing server executes the monitoring script transmitted from the management server 1, confirms the operating state of the service providing server, and returns the confirmation result to the management server 1. Then, the server monitoring processing unit 3 monitors the operating state of the service providing server based on the returned execution result.

When the execution result of the monitoring script indicates an abnormality of the service providing server, the cluster control processing unit 4 requests the service providing server to execute the control script for performing failover. The cluster control processing unit 4 is not limited to failover control as a series of recovery operations including start and stop of services in an abnormal situation, and may perform other controls. For example, the cluster control processing unit 4 may perform service start control or service migration (service start and stop) control in a normal state.

In the HA cluster system including the management server 1 and the service providing server described above, the management server 1 determines whether or not the service providing server is in an abnormal state, and the management server 1 controls the execution of failover. That is, the management server 1 monitors the OS function provision status and the service function provision status of the service providing server via the network, and when a failure is detected, the management server 1 can control the recovery operation of the system. .. As described above, the management server 1 is provided with the clusterware, and the service providing server does not have to have the clusterware. As a result, the HA cluster system can be provided in an agentless manner.

In an HA cluster system that does not have the management server 1 and has clusterware on the service providing server (hereinafter referred to as the cluster system according to the comparative example), the service providing server is restarted and the system is restarted by installing the clusterware and constructing the cluster. Will stop. Therefore, there is a tendency to build a completely new cluster environment and then move to a new system without utilizing the existing environment.

On the other hand, in the case of agentless clusterware, it is not necessary to install the clusterware for each managed server (that is, the service providing server), and it is sufficient to install it only on the management server 1. Therefore, it does not affect the existing environment such as system stoppage. And it is not necessary to build a completely new environment, and the existing environment can be effectively utilized. Even if the clusterware version upgrade or maintenance is required, there is no need to stop the system or install each managed server. As a result, it is possible to reduce the introduction and operation costs. Therefore, maintenance and system expansion can be easily performed.

In addition, in the cluster system of the comparative example, the survival confirmation process between the service providing servers and the alive monitoring process for monitoring the monitoring process are performed periodically, so there is a risk that the service providing server will be overloaded. is there. Therefore, if the service providing server has few resources, monitoring may fail, and it may be erroneously determined that a failure has occurred.

On the other hand, in the HA cluster system having the management server 1 described above, since the main processing for achieving high availability is performed by the management server 1, the load on the service providing server (that is, the cluster server) is suppressed. be able to.

Next, the details of the embodiment will be described.
<Embodiment 1>
FIG. 2 is a block diagram showing an example of the HA cluster system 10 according to the first embodiment. Further, FIG. 3 is a block diagram showing an example of the hardware configuration of the operation management server 100.

As shown in FIG. 2, the HA cluster system 10 includes an operation management server 100, a working business server 200, and a standby business server 300. The working business server 200 and the standby business server 300 are servers for providing predetermined business services, and the operation management server 100 controls failover and the like in the working business server 200 and the standby business server 300. It is a server.

The operation management server 100, the active business server 200, and the standby business server 300 are connected to the network 400. Further, the operation management server 100, the active business server 200, and the standby business server 300 are connected to each other by the network 401.

The operation management server 100 corresponds to the management server 1 of FIG. 1, and the active business server 200 and the standby business server 300 correspond to the above-mentioned service providing server. In the HA cluster system 10, the active business server 200 and the standby business server 300 are cluster servers, and the active system and the standby system are switched by failover. That is, due to failover, the active business server 200 is switched to the standby system, and the standby business server 300 is switched to the active system.
In the following description, the active business server 200 and the standby business server 300 may be referred to as managed servers.

The network 400 is a public LAN (Local Area Network) that can be connected to a client device (not shown). The working business server 200 provides business services to the client device via the network 400. The network 400 is also used to monitor the provision status of business services by the working business server 200 from the operation management server 100.
The network 401 is an interconnect LAN that interconnects the operation management server 100, the active business server 200, and the standby business server 300. The network 401 is used for internal communication (communication used for server monitoring, business service control, cluster control, etc.) between the operation management server 100 and the active business server 200 or the standby business server 300.

As shown in FIG. 2, in the HA cluster system 10, the clusterware 110 is provided on the operation management server 100. The clusterware 110 includes a cluster control unit 111, a business service control unit 112, a server monitoring unit 113, an internal monitoring unit 114, a script execution unit 115, a server communication unit 116, and a script storage unit 117.

The cluster control unit 111 performs various processes for controlling the HA cluster system 10. That is, the cluster control unit 111 controls, for example, the execution of failover.
The business service control unit 112 controls the start and stop of the business service providing unit 201 of the working business server 200 and the business service providing unit 301 of the standby business server 300.

The server monitoring unit 113 monitors the operating status of the active business server 200 and the standby business server 300. For example, the server monitoring unit 113 determines whether or not the business service is normally provided in the working business server 200, and no abnormality such as hardware or software has occurred in the working business server 200 and the standby business server 300. Monitor.

The internal monitoring unit 114 (also referred to as an internal monitoring processing unit) monitors the operating states of the cluster control unit 111, the business service control unit 112, and the server monitoring unit 113. It is preferable that the internal monitoring unit 114 monitors all of the cluster control unit 111, the business service control unit 112, and the server monitoring unit 113, but the operating state of only a part of these may be monitored. The operating status of other units may be monitored.
Further, when the internal monitoring unit 114 detects that these operating states are abnormal, the internal monitoring unit 114 may notify another device, an operation manager, or the like of the occurrence of the abnormality. As a result, it is possible to execute a predetermined process when an abnormality occurs in the HA cluster system 10 or to perform a predetermined operation by the operation manager.
Specifically, the internal monitoring unit 114 performs alive monitoring of the processes of the cluster control unit 111, the business service control unit 112, and the server monitoring unit 113.

The script execution unit 115 executes the script stored in the script storage unit 117. For example, the script execution unit 115 executes a script for monitoring the active business server 200 and the standby business server 300 at the request of the server monitoring unit 113. Further, for example, the script execution unit 115 executes a script for starting or stopping the business service providing units 201 and 301 in the active business server 200 or the standby business server 300 at the request of the business service control unit 112. .. In addition, the script execution unit 115 generates a script from the setting file, and performs processing such as encrypting the generated script with a predetermined key.

The server communication unit 116 performs communication establishment processing and disconnection processing with other servers.
The script storage unit 117 is a storage area for storing a script and a setting file or the like that is a basis for generating the script.

The working business server 200 has a business service providing unit 201, a script execution unit 202, and a script storage unit 203. Similarly, the standby business server 300 also has a business service providing unit 301, a script execution unit 302, and a script storage unit 303.

The business service providing units 201 and 301 provide predetermined business services to a client device (not shown) connected to the network 400 via the network 400. That is, the business service providing units 201 and 301 are application programs that provide a predetermined business service.

The script execution unit 202 executes the script stored in the script storage unit 203 in response to a request from the operation management server 100. Similarly, the script execution unit 302 executes the script stored in the script storage unit 303 in response to a request from the operation management server 100. For example, the script execution unit 202 executes a script for checking the operating state of the active business server 200 and a script for starting or stopping the business service providing unit 201. Similarly, the script execution unit 302 executes a script for checking the operating state of the standby business server 300 and a script for starting or stopping the business service providing unit 301. The script execution units 202 and 302 are not resident processes, but are temporarily created in response to a request from the operation management server 100, and execute the scripts of the script storage units 203 and 303.
The script storage units 203 and 303 are storage areas for storing the script received from the operation management server 100.

As shown in FIG. 3, the operation management server 100 includes, for example, a network interface 150, a memory 151, and a processor 152. In addition to this, the operation management server 100 may include a storage device such as a hard disk drive.

The network interface 150 is used to perform communication via the networks 400 and 401. The network interface 150 may include, for example, a network interface card (NIC).

The memory 151 is composed of a combination of a volatile memory and a non-volatile memory. Memory 151 may include storage located away from processor 152. In this case, the processor 152 may access the memory 151 via an input / output interface (not shown).

The memory 151 is used, for example, to store software (computer program) including one or more instructions executed by the processor 152.

This program can be stored and supplied to a computer using various types of non-transitory computer readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-temporary computer-readable media include magnetic recording media (eg flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (eg magneto-optical disks), Compact Disc Read Only Memory (CD-ROM), CD- Includes R, CD-R / W, semiconductor memory (eg, mask ROM, Programmable ROM (PROM), Erasable PROM (EPROM), flash ROM, Random Access Memory (RAM)). The program may also be supplied to the computer by various types of transient computer readable media. Examples of temporary computer-readable media include electrical, optical, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

The processor 152 may be, for example, a microprocessor, an MPU (Micro Processor Unit), or a CPU (Central Processing Unit). The processor 152 may include a plurality of processors. The processor 152 reads a computer program from the memory 151 and executes it to process the cluster control unit 111, the business service control unit 112, the server monitoring unit 113, the internal monitoring unit 114, the script execution unit 115, and the server communication unit 116. I do. Further, the script storage unit 117 is realized by, for example, a memory 151.

The working business server 200 and the standby business server 300 also have the same hardware configuration as that shown in FIG. Therefore, the processor of the working business server 200 reads the computer program from the memory and executes it to perform the processing of the business service providing unit 201 and the script executing unit 202. Further, the script storage unit 203 is realized by, for example, the memory of the working business server 200. Similarly, the processor of the standby business server 300 reads the computer program from the memory and executes it to perform the processing of the business service providing unit 301 and the script executing unit 302. Further, the script storage unit 303 is realized by, for example, the memory of the standby business server 300.

Hereinafter, the operation of each component of the HA cluster system 10 will be described with reference to the flowcharts of FIGS. 4 to 8.

FIG. 4 is a flowchart showing an operation flow of the cluster construction process in the HA cluster system 10. By this construction process, a script is placed on each server. Hereinafter, the operation flow will be described with reference to FIG.

In step 100 (S100), the operation manager inputs setting information related to monitoring required for cluster construction via the GUI (Graphical User Interface) of the operation management server 100. Specifically, the operation administrator can monitor targets (server name, business service, disk, processor, etc.), monitoring settings (interval, timeout, number of retries, etc.), and what to do when a failure occurs (process restart, failover, etc.) , Server restart, etc.), etc.

Next, in step 101 (S101), when the cluster control unit 111 receives the input from the GUI, the cluster control unit 111 requests the script execution unit 115 to create a setting file according to the input contents. Then, in step 102 (S102), the script execution unit 115 creates a setting file based on the input contents by executing a predetermined script for generating the setting file, and stores the setting file in the script storage unit 117.

Next, in step 103 (S103), the cluster control unit 111 requests the script execution unit 115 to generate a script for executing the process according to the setting file. Then, the script execution unit 115 executes a predetermined script for script generation. As a result, in step 104 (S104), the script execution unit 115 acquires the setting file from the script storage unit 117, and in step 105 (S105), generates a script capable of executing the process according to the setting file. The script execution unit 115 stores the generated script in the script storage unit 117.

Next, in step 106 (S106), the cluster control unit 111 requests the script execution unit 115 to transfer the script for the managed server generated in step 105. The script execution unit 115 transfers the script to the management target server (active business server 200 and standby business server 300) as follows by executing a predetermined script for transfer.
In step 107 (S107), the script execution unit 115 acquires the script for the managed server from the script storage unit 117, and in step 108 (S108), encrypts the script using a predetermined private key. Then, in step 109 (S109), the script execution unit 115 transfers the encrypted script to the managed server.

The active business server 200 stores the received script in the script storage unit 203, and the standby business server 300 stores the received script in the script storage unit 303. Therefore, in the present embodiment, it is not necessary to transfer the script each time the script is executed on the managed server, and the time required for executing the script can be suppressed. In addition, since the encrypted script is stored in the managed server, security can be ensured.

As described above, the cluster control unit 111 and the script execution unit 115 correspond to the transmission processing unit 2 of FIG. That is, the cluster control unit 111 and the script execution unit 115 perform a process of transmitting a monitoring script for monitoring the operation of the managed server (state of the managed server) to the managed server. Further, the cluster control unit 111 and the script execution unit 115 provide a control script for controlling the operation of the managed server (for example, start / stop of the business service providing units 201 and 301) regarding failover to the managed server. Performs the process of sending to.

When changing the cluster settings, the cluster control unit 111 requests the script execution unit 115 to update the setting file based on the changed contents. On the other hand, the script execution unit 115 performs the same processing as described above, and transfers the encrypted script to the managed server again.

FIG. 5 is a flowchart showing a flow of operation for starting a cluster in the HA cluster system 10. Hereinafter, the operation flow will be described with reference to FIG.

In step 200 (S200), the internal monitoring unit 114 starts monitoring the operating state of the cluster control unit 111. That is, the internal monitoring unit 114 starts alive monitoring of the process of the cluster control unit 111.

Next, in step 201 (S201), the cluster control unit 111 activates the business service control unit 112. That is, the cluster control unit 111 starts the process of the business service control unit 112.
When the business service control unit 112 is activated, in step 202 (S202), the internal monitoring unit 114 starts monitoring the operating state of the business service control unit 112. That is, the internal monitoring unit 114 starts the life-and-death monitoring of the process of the business service control unit 112.

Next, in step 203 (S203), the cluster control unit 111 requests the business service control unit 112 to start the business service providing unit 201 in the active business server 200.
In step 204 (S204), the business service control unit 112 requests the script execution unit 115 to start the business service providing unit 201 in the active business server 200.
In step 205 (S205), the script execution unit 115 requests the active business server 200 to start the business service providing unit 201 by executing the script that requests the active business server 200 to start.

In step 206 (S206), the script execution unit 202 of the active business server 200 that received the request from the operation management server 100 acquires the script for starting the business service providing unit 201 from the script storage unit 203. The script for starting the business service providing unit 201 is stored in the script storage unit 203 in advance by the process shown in FIG.
In step 207 (S207), the script execution unit 202 of the active business server 200 starts the business service providing unit 201 by executing the acquired script. As a result, the provision of the business service from the active business server 200 to the client device (user) via the network 400 is started (step 208 (S208)).

In step 209 (S209), the cluster control unit 111 of the operation management server 100 that has requested the start of the business service providing unit 201 starts the server monitoring unit 113. That is, the cluster control unit 111 starts the process of the server monitoring unit 113.
When the server monitoring unit 113 is activated, in step 210 (S210), the internal monitoring unit 114 starts monitoring the operating state of the server monitoring unit 113. That is, the internal monitoring unit 114 starts alive monitoring of the process of the server monitoring unit 113.

FIG. 6 is a flowchart showing the flow of the monitoring operation in the HA cluster system 10. Hereinafter, the operation flow will be described with reference to FIG. Since the monitoring is continuously performed, the operation shown in FIG. 6 is repeatedly executed.

In step 300 (S300), the server monitoring unit 113 requests the script execution unit 115 to perform a process for confirming the provision status of the business service by the active business server 200. Specifically, the script execution unit 115 is requested to perform a process of confirming whether or not the business service provided by the business service providing unit 201 of the working business server 200 can be accessed from the operation management server 100 via the network 400. ..
In step 301 (S301), the script execution unit 115 confirms the access path to the business service by executing a predetermined script for confirming the provision status of the business service. As a result, it is confirmed whether or not the business service of the working business server 200 can be normally accessed via the network 400.

Further, in the HA cluster system 10, the monitoring script is executed in the active business server 200 and the standby business server 300 in response to the request from the operation management server 100 via the network 401.
Specifically, first, in step 302 (S302), the server monitoring unit 113 requests the script execution unit 115 to confirm the operating status of the managed server (active business server 200 and standby business server 300).
In step 303 (S303), the script execution unit 115 executes a predetermined script for requesting operation confirmation, thereby causing the managed server (active business server 200 and standby business server 300) to operate. Request confirmation of. That is, the script execution unit 115 requests the active business server 200 to execute the script for monitoring the active business server 200, and the standby business server 300 is requested to execute the script for monitoring the standby business server 300. Request the execution of.

In step 304 (S304), the managed server (active business server 200 and standby business server 300) confirms the status of its own device by executing a monitoring script. That is, the script execution unit 202 of the active business server 200 that receives the request from the operation management server 100 acquires the monitoring script from the script storage unit 203. Then, the script execution unit 202 of the working business server 200 confirms the operating state of the working business server 200 by executing the acquired script. Similarly, the script execution unit 302 of the standby business server 300 that receives the request from the operation management server 100 acquires the monitoring script from the script storage unit 303. Then, the script execution unit 302 of the standby business server 300 confirms the operating state of the standby business server 300 by executing the acquired script. The monitoring script is stored in the script storage units 203 and 303 in advance by the process shown in FIG.
By executing the monitoring script, the status information of the hardware and software of the managed server is acquired. Specifically, for example, status information such as the life-and-death status of the application process, the operating status of the access path to the disk, the link status of the LAN cable, and the NIC status is acquired. The script execution units 202 and 302 transmit the execution result of the monitoring script, that is, the status information of the managed server to the operation management server 100, and the script execution unit 115 of the operation management server 100 acquires this (step 305 (step 305 (step 305)). S305)).

In step 306 (S306), the script execution unit 115 outputs the result obtained in step 301 and the result obtained in step 305 to the server monitoring unit 113 as monitoring information.
Next, in step 307 (S307), the server monitoring unit 113 determines whether or not the managed server is normal, that is, whether or not a failure has occurred, based on the monitoring information acquired in step 306. judge. The determination in step 307 may be performed by analyzing the monitoring information (for example, comparing the monitoring result expected when the managed server is normal with the actual monitoring result).

As described above, the server monitoring unit 113 and the script execution unit 115 correspond to the server monitoring processing unit 3 of FIG. That is, the server monitoring unit 113 and the script execution unit 115 request the managed server to execute the monitoring script and return the execution result, and monitor the operating state of the managed server based on the execution result.

FIG. 7 is a flowchart showing a failover operation in the HA cluster system 10. When a failure occurs in the active business server 200, the HA cluster system 10 performs the following processing. That is, when a failure of the active business server 200 is detected in step 307 of FIG. 6, the HA cluster system 10 performs the following processing. Hereinafter, the operation flow will be described with reference to FIG. 7.

In step 400 (S400), the server monitoring unit 113 that detects that a failure has occurred in the active business server 200 notifies the cluster control unit 111 of the occurrence of an abnormality.
Next, in step 401 (S401), the cluster control unit 111, which has been notified of the occurrence of the abnormality, requests the server monitoring unit 113 to confirm whether the standby business server 300 can normally take over the business service. The server monitoring unit 113 determines whether or not the standby business server 300 can normally take over the business service based on the state of the standby business server 300 obtained in step 307 of FIG. When the standby business server 300 is normal, the server monitoring unit 113 determines that the standby business server 300 can take over the business service normally.

When the standby business server 300 can take over the business service normally, the cluster control unit 111 requests the business service control unit 112 to fail over. Specifically, the following processing is performed, and the state of the system is switched from the provision of the business service by the active business server 200 to the provision of the business service by the standby business server 300.

In step 402 (S402), the cluster control unit 111 requests the business service control unit 112 to control to stop the provision of the business service by the active business server 200. On the other hand, in step 403 (S403), the business service control unit 112 requests the script execution unit 115 to stop the business service providing unit 201 in the active business server 200.
In step 404 (S404), the script execution unit 115 requests the working business server 200 to stop the business service providing unit 201 by executing the script that requests the working server 200 to stop. That is, the script execution unit 115 requests the active business server 200 to execute the script for stopping the business service providing unit 201.
Then, in step 405 (S405), the script execution unit 202 of the working business server 200 that received the request executes the script for stopping the business service providing unit 201. That is, the script execution unit 202 of the active business server 200 that receives the request from the operation management server 100 acquires the script for stopping the business service providing unit 201 from the script storage unit 203. Then, the script execution unit 202 of the active business server 200 stops the provision of the business service by the active business server 200 by executing the acquired script. The stop script is stored in advance in the script storage unit 203 by the process shown in FIG.

Further, in step 406 (S406), the cluster control unit 111 requests the business service control unit 112 to control the standby system business server 300 to start providing the business service. On the other hand, in step 407 (S407), the business service control unit 112 requests the script execution unit 115 to start the business service providing unit 301 in the standby system business server 300.
In step 408 (S408), the script execution unit 115 requests the standby business server 300 to start the business service providing unit 301 by executing the script that requests the standby business server 300 to start. That is, the script execution unit 115 requests the standby business server 300 to execute the script for starting the business service providing unit 301.
Then, in step 409 (S409), the script execution unit 302 of the standby business server 300 that received the request executes the script for starting the business service providing unit 301. That is, the script execution unit 302 of the standby business server 300 that receives the request from the operation management server 100 acquires the script for starting the business service providing unit 301 from the script storage unit 303. Then, the script execution unit 302 of the standby business server 300 starts providing the business service by the standby business server 300 by executing the acquired script. The activation script is stored in advance in the script storage unit 303 by the process shown in FIG.
In this way, the business service can be provided from the standby business server 300, and the failover is completed.

As described above, the cluster control unit 111, the business service control unit 112, and the script execution unit 115 correspond to the cluster control processing unit 4 in FIG. That is, the cluster control unit 111, the business service control unit 112, and the script execution unit 115 request the managed server to execute a script for failover when the active business server 200 is abnormal.

FIG. 8 is a flowchart showing an operation related to script execution in the HA cluster system 10. In FIG. 8, the upper row shows the flow of operations when the operation management server 100 causes the managed server to execute the script, and the lower row shows the flow of operations when the operation management server 100 executes the script. Hereinafter, the operation flow will be described with reference to FIG.

First, the flow of the operation (remote execution operation) when the operation management server 100 causes the managed server (active business server 200 and standby business server 300) to execute the script will be described.
In step 500 (S500), the script execution unit 115 of the operation management server 100 requests the server communication unit 116 to establish a communication session with the managed server. On the other hand, in step 501 (S501), the server monitoring unit 113 establishes a communication session to the managed server, that is, the active business server 200 or the standby business server 300. When the process according to step 301 in FIG. 6 is performed, the server communication unit 116 establishes a communication session on the network 400 (public LAN). In other cases, the server communication unit 116 establishes a communication session on the network 401 (interconnect LAN).

Next, in step 502 (S502), the script execution unit 115 requests the script execution unit 202 or 302 of the managed server to execute the script in accordance with the request from the business service control unit 112 or the server monitoring unit 113. .. At this time, the script execution unit 115 transmits the key required for decrypting the script encrypted and stored in the script storage units 203 and 303.

In step 503 (S503), the script execution units 202 and 302 acquire the script from the script storage units 203 and 303. Then, in step 504 (S504), the script is decrypted using the received key, and in step 505 (S505), the decrypted script is executed. After that, in step 506 (S506), the script execution units 202 and 302 transmit the script execution result to the script execution unit 115 of the operation management server 100.

Next, in step 507 (S507), the script execution unit 115 that has acquired the execution result from the managed server requests the server communication unit 116 to disconnect the communication session to the managed server. On the other hand, in step 508 (S508), the server monitoring unit 113 disconnects the communication session to the managed server. The script execution unit 115 that has acquired the execution result outputs the execution result to the business service control unit 112 or the server monitoring unit 113 that is the request source.
As described above, in the present embodiment, the script is executed by remote login from the operation management server 100 to the managed server. In remote login, SSH (Secure Shell) may be used.

Next, the flow of the operation (local execution operation) when the script is executed on the operation management server 100 will be described. In this case, as shown in the lower part of FIG. 8, processing such as communication session establishment, disconnection, and script decoding is not performed, and the script in the script storage unit 117 is executed. That is, in step 550 (S550), the script execution unit 115 acquires the script to be executed from the script storage unit 117. Then, in step 551 (S551), the script execution unit 115 executes the acquired script. By such processing, process monitoring (monitoring by the internal monitoring unit 114) of the clusterware 110 on the operation management server 100 is executed. In this case, the execution result of the script execution unit 115 is output to the internal monitoring unit 114 of the request source.

The first embodiment has been described above. In the HA cluster system 10 according to the first embodiment, the operation management server 100 is provided with the above-mentioned clusterware 110, and the managed server executes a script and returns an execution result in response to a request from the clusterware 110. Do. Therefore, it is possible to avoid the occurrence of problems caused by the managed server having clusterware (for example, problems such as the inability to effectively utilize the existing environment and the processing load of the clusterware).

Further, in the HA cluster system 10, as described with reference to FIG. 6, the operation management server 100 is operated by the managed server via the network 400 used to provide the business service to the client device. Monitor the service provision status. Then, when the operation management server 100 detects that the business service is not normally provided by the managed server by such monitoring, the operation management server 100 executes a script for controlling the operation of the managed server regarding failover. To the managed server.
By the way, when it is the server of the provider that determines whether or not the business service can be provided, it cannot be reliably determined whether or not the business service can be normally provided to the external client device via the network 400.
On the other hand, since the HA cluster system 10 is accessed in the same manner as an actual external client device, it is possible to more accurately determine whether or not the business service can be provided normally.

<Embodiment 2>
Next, the second embodiment will be described. In the following description, the description of the configuration and operation overlapping with the first embodiment will be omitted. This embodiment is different from the above-described embodiment in that the operation management server performs processing on the BMC (Baseboard Management Controller) included in the managed server.

FIG. 9 is a block diagram showing an example of the HA cluster system 20 according to the second embodiment. As shown in FIG. 9, the working business server 200 includes a BMC 204, and similarly, the standby business server 300 includes a BMC 304. The BMCs 204 and 304 operate independently of the OS to monitor the hardware that constitutes the server, such as a processor and memory, and to control the start and stop of the server. The BMCs 204 and 304 conform to, for example, an IPMI (Intelligent Platform Management Interface) standard, and are connected to a network 402 which is a LAN (BMC LAN) for communication with the BMC. The BMCs 204 and 304 provide the operation management server 100 with a failure reporting function and a remote control function even when the OS of the managed server is down due to a hardware failure or the like.
When the OS of the managed server goes down, it becomes impossible to monitor and control the managed server by causing the managed server to execute a script from the operation management server 100 via the network 401. Even in such a case, in the present embodiment in which the BMC is mounted on the managed server, the managed server is monitored and controlled from the operation management server 100 via the network 402 as shown in FIG. Life and death monitoring, forced stop, etc.) can be performed.

FIG. 10 is a flowchart showing the operation of server monitoring and server control using BMC in the HA cluster system 20. Hereinafter, the operation flow will be described with reference to FIG.

In step 600 (S600), the server monitoring unit 113 requests the script execution unit 115 to confirm the operating status of the managed servers (active business server 200 and standby business server 300).
In step 601 (S601), the script execution unit 115 executes a predetermined script for requesting the status information of the managed server. That is, the script execution unit 115 transmits a command (IPMI command for monitoring) for acquiring the operating state of the managed server from the BMCs 204 and 304 via the network 402.

In step 602 (S602), the BMCs 204 and 304 that have received the command acquire monitoring information (also referred to as status information) that is various hardware information (CPU information, memory information, OS status, etc.).
Then, in step 603 (S603), the BMCs 204 and 304 transmit the monitoring information to the operation management server 100.
In step 604 (S604), the script execution unit 115 of the operation management server 100 that has acquired the monitoring information from the BMCs 204 and 304 outputs the monitoring information to the server monitoring unit 113.
In step 605 (S605), the server monitoring unit 113 analyzes the monitoring information acquired in step 604 and confirms the status of the managed server. That is, the server monitoring unit 113 determines whether or not the managed server is normal, that is, whether or not a failure has occurred.

When the server monitoring unit 113 detects an abnormality in the managed server in step 605, the server monitoring unit 113 notifies the cluster control unit 111 of the occurrence of the abnormality in the managed server in step 606 (S606).
In step 607 (S607), the cluster control unit 111 that has received the notification requests the script execution unit 115 to perform processing (restart of the managed server, etc.) when a failure occurs.
In step 608 (S608), the script execution unit 115 executes a predetermined script for requesting the BMCs 204 and 304 to perform processing when a failure occurs. That is, the script execution unit 115 transmits, for example, a power-off command or a power-on command (IPMI command for power control) of the managed server to the BMCs 204 and 304 via the network 402.
In step 609 (S609), the BMCs 204 and 304 that have received the command control the power supply of the managed server. As a result, for example, as a process when a failure occurs, the managed server in which the failure has occurred is restarted. Alternatively, for example, as a process when a failure occurs, the active business server 200 in which the failure has occurred is stopped and the standby business server 300 is started. If a failure of the working business server 200 is detected while the standby business server 300 is stopped, the power of the working business server 200 can be turned off and the power of the standby business server 300 can be turned on. Internet Protocol) Address conflicts do not occur. That is, it is not necessary to switch the connection from the client device.

When the processing by BMC 204 and 304 in step 609 is performed, the execution result of the processing is returned to the script execution unit 115 of the operation management server 100 in step 610 (S610). Further, in step 611 (S611), the cluster control unit 111 acquires the execution result from the script execution unit 115.

As described above, in the present embodiment, the operation management server 100 requests the BMC of the managed server to return the status information of the managed server, and monitors the status of the managed server based on this status information. .. Then, when the operation management server 100 obtains the state information indicating the abnormality of the managed server from the BMC, the operation management server 100 requests the BMC to control the power supply of the managed server. According to such a configuration, it is possible to appropriately deal with the situation where the monitoring or control script cannot be executed on the managed server.

<Embodiment 3>
Next, the third embodiment will be described. This embodiment is different from the first embodiment in that any of the servers that provide business services is also used as an operation management server. Hereinafter, the points different from those of the first embodiment will be described, and the description of the same configuration and operation as that of the first embodiment will be omitted.

FIG. 11 is a block diagram showing an example of the HA cluster system 30 according to the third embodiment. As shown in FIG. 11, the HA cluster system 30 has an operation management / standby business server 500 and a working business server 200. As shown in FIG. 11, the operation management / standby business server 500 is different from the operation management server 100 according to the first embodiment in that it has a business service providing unit 120 in addition to the clusterware 110. That is, the operation management / standby business server 500 can be said to be a server to which the business service providing unit 120 is added to the operation management server 100 according to the first embodiment. The business service providing unit 120 performs the same operation as the above-mentioned business service providing units 201 and 301.

In the present embodiment, when the operation management / standby business server 500 monitors its own server, the server monitoring unit 113 sends a script to the script execution unit 115 in the same manner as the operation shown in the lower part of FIG. Requests that the script in storage 117 be executed locally. Further, when starting or stopping the business service providing unit 120, the script executing unit 115 is requested to execute the script of the script storage unit 117 locally, as in the operation shown in the lower part of FIG.

As described above, since the server that performs operation management has the business service providing unit 120, the server can be used not only as operation management but also as a standby server. Further, according to such a configuration, the number of servers prepared as standby servers can be reduced, so that the system introduction cost can be suppressed.

The present invention is not limited to the above embodiment, and can be appropriately modified without departing from the spirit. For example, in the third embodiment, the control using the BMC described in the second embodiment may be performed.

Although the invention of the present application has been described above with reference to the embodiments, the invention of the present application is not limited to the above. Various changes that can be understood by those skilled in the art can be made within the scope of the invention in the configuration and details of the invention of the present application.

This application claims priority on the basis of Japanese application Japanese Patent Application No. 2018-39390 filed on March 6, 2018, and the entire disclosure thereof is incorporated herein by reference.

1 Management server 2 Transmission processing unit 3 Server monitoring processing unit 4 Cluster control processing unit 10, 20, 30 HA cluster system 100 Operation management server 110 Clusterware 111 Cluster control unit 112 Business service control unit 113 Server monitoring unit 114 Internal monitoring unit 115 , 202, 302 Script execution unit 116 Server communication unit 117, 203, 303 Script storage unit 120, 201, 301 Business service provision unit 150 Network interface 151 Memory 152 Processor 200 Active business server 300 Standby business server 400, 401, 402 Network 500 Operation management / standby business server

Claims (9)

Performs a process of transmitting a monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server regarding failover to the service providing server. Transmission processing means and
A server monitoring processing means that requests the service providing server to execute the monitoring script and return the execution result, and monitors the operating state of the service providing server based on the execution result.
A management server having a cluster control processing means that requests the service providing server to execute the control script when the execution result of the monitoring script indicates an abnormality of the service providing server. The server monitoring processing means further monitors the provision status of the predetermined service by the service providing server via the network used by the service providing server to provide the predetermined service to the client device. And
The management server according to claim 1, wherein the cluster control processing means further requests the service providing server to execute the control script when the predetermined service is not normally provided by the service providing server. The service providing server is a server provided with a BMC (Baseboard Management Controller).
The server monitoring processing means further requests the BMC of the service providing server to return the state information of the service providing server, and monitors the state of the service providing server based on the state information.
The management server according to claim 1 or 2, wherein the cluster control processing means requests the BMC to control the power supply of the service providing server when state information indicating an abnormality of the service providing server is obtained from the BMC. .. The management server according to any one of claims 1 to 3, further comprising an internal monitoring processing means for monitoring the operating state of at least one of the server monitoring processing means or the cluster control processing means. The management server according to any one of claims 1 to 4, further comprising a service providing means for providing the predetermined service. A working server for providing a given service and
A standby server for providing the prescribed service and
The management server that controls failover in the active server and the standby server,
With
The management server
A process of transmitting a first monitoring script for monitoring the operation of the active server and a first control script for controlling the operation of the active server regarding failover to the active server. A process of transmitting a second monitoring script for monitoring the operation of the standby server and a second control script for controlling the operation of the standby server regarding failover to the standby server. Transmission processing means to be performed and
A server monitoring process that requests the active server and the standby server to execute the monitoring script and return the execution result, and monitors the operating status of the active server and the standby server based on the execution result. Means and
When the execution result of the monitoring script of the working server indicates an abnormality of the working server, the working server is requested to execute the first control script and the second control script is executed. A cluster system that has the cluster control processing means required for the standby server. The server monitoring processing means further monitors the provision status of the predetermined service by the active server via the network used by the active server to provide the predetermined service to the client device. And
The cluster control processing means further requests the active server to execute the first control script and the second control script when the predetermined service is not normally provided by the active server. The cluster system according to claim 6, which requests the standby server to execute the above. A monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server regarding failover are transmitted to the service providing server.
The service providing server is requested to execute the monitoring script and return the execution result, and the operating state of the service providing server is monitored based on the execution result.
A method of controlling a cluster system that requests the service providing server to execute the control script when the execution result of the monitoring script indicates an abnormality of the service providing server. Performs a process of transmitting a monitoring script for monitoring the operation of the service providing server that provides a predetermined service and a control script for controlling the operation of the service providing server regarding failover to the service providing server. Transmission processing step and
A server monitoring processing step that requests the service providing server to execute the monitoring script and return the execution result, and monitors the operating state of the service providing server based on the execution result.
When the execution result of the monitoring script indicates an abnormality of the service providing server, a non-temporary program containing a program for causing the computer to execute the cluster control processing step requesting the service providing server to execute the control script is stored. Computer readable medium.

Images