JP2016148973A - Life-and-death monitoring device, life-and-death monitoring system, life-and-death monitoring method, and life-and-death monitoring method program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a life-and-death monitoring device for performing life-and-death monitoring of a multi server system for providing a service by combining plural servers having different functions each other.SOLUTION: A life-and-death monitoring device 30 comprises: a life-and-death monitoring request sending part 31 for requesting life-and-death monitoring of a multi-server system 100 which provides a service by combining plural servers having different functions each other, by sending a service request signal of the multi-server system; and a life-and-death monitoring part 32. The life-and-death monitoring request sending part 31 requests the life-and-death monitoring for monitoring normality of the multi-server system 100 by sending a normal service request signal which the system uses, to the multi-server system 100. The life-and-death monitoring part 32 performs the life-and-death monitoring based on a status code of a response signal to the service request signal.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a life and death monitoring apparatus, a life and death monitoring system, a life and death monitoring method, and a life and death monitoring method program for monitoring the life and death of a plurality of servers constituting the system.

  In a system that provides services by combining multiple servers with different functions, managing the performance and normality of all servers used to provide services is important in maintaining service provision and downtime. It is important for minimization.

  A server alive monitoring method is known in which a normal or failure of an application server (hereinafter referred to as a server) is confirmed by sending a signal from another server (Non-Patent Document 1). A conventional alive monitoring method transmits a alive monitoring signal to each server from the outside connected via a network. The server that has received the alive monitoring signal determines whether or not the server is operating correctly, and returns the determination result to monitor the server alive activity.

http://asp.ag-monitor.com/about_watching.php

  A conventional alive monitoring method is a method in which a server that has received a alive monitoring signal diagnoses only the operation of an application of its own server, and is used in a system including a plurality of servers having equivalent functions. The feature is to judge only the normality of the operation of the application of the own server.

  Therefore, it cannot be used for a system that provides a service by combining servers having different functions. That is, the conventional method is a method of responding to the server that monitors life and death with the result of determining that each server is normal, and only knows whether each server is normal or malfunctioning.

  Therefore, there is a problem that normality including cooperation between servers of a multi-server system that provides a service by combining a plurality of servers having different functions cannot be monitored.

  The present invention has been made in view of this problem, and is a life / death monitoring apparatus, a life / death monitoring system, a life / death monitoring method for performing life / death monitoring of a multi-server system that provides a service by combining servers having a plurality of different functions, It is another object of the present invention to provide a life and death monitoring method program.

  The life and death monitoring apparatus of the present invention uses a service request signal provided by the multi-server system as a signal for life and death monitoring of a multi-server system that provides a service by combining a plurality of servers having different functions. A life / death monitoring request transmission unit for transmitting to any one of them, a response signal for the request signal, and a status of the transmission destination server of the request signal included in the response signal and a status of the cooperation destination server in cooperation with the transmission destination server And a life / death monitoring unit that performs life / death monitoring of the plurality of servers.

  In addition, the life and death monitoring system of the present invention is a life and death monitoring system comprising a multi-server system that provides a service by combining a plurality of servers having different functions, and a life and death monitoring device, As a life and death monitoring signal of the multi-server system, a life and death monitoring request transmission unit that transmits a request signal of a service provided by the multi-server system to any of the plurality of servers, and a response signal to the request signal, Using a status of the destination server of the request signal included in the response signal, a status of the cooperation destination server in cooperation with the destination server, and failure information, and a life / death monitoring unit that performs life / death monitoring of the plurality of servers And the server adds the failure information to the response signal for the service request signal. And summarized in that it includes the failure information adding unit that.

  In addition, the life and death monitoring method of the present invention is a life and death monitoring method performed by the life and death monitoring device, and the multi-server is used as a life and death monitoring signal of a multi-server system that provides a service by combining a plurality of servers having different functions. A life / death monitoring request transmission step for transmitting a request signal of a service provided by the system to any of the plurality of servers, a response signal for the request signal, and a transmission destination server of the request signal included in the response signal The gist is to perform a life and death monitoring step of performing life and death monitoring of the plurality of servers using the status and the status of the cooperation destination server that cooperates with the transmission destination server.

  Another life and death monitoring method of the present invention is a life and death monitoring method performed by a life and death monitoring system comprising a multi-server system that provides a service by combining a plurality of servers having different functions, and a life and death monitoring device, A failure in which the life and death monitoring device performs a life and death monitoring request transmission step of transmitting a request signal for the service as a life and death monitoring signal of the multi-server system, and the server adds failure information to a response signal to the request signal Performing an information addition step, wherein the alive monitoring device receives a response signal to the request signal, and the status of the transmission destination server of the request signal included in the response signal and the status of the cooperation destination server that cooperates with the transmission destination server A life / death monitoring step for performing life / death monitoring of a plurality of the servers using the failure information The gist that you do.

  The life and death monitoring method program of the present invention is a program for causing a computer or a computer system to execute the above life and death monitoring method.

  According to the life and death monitoring apparatus and life and death monitoring method of the present invention, it is possible to monitor the normality of the cooperative operation between servers of a multi-server system providing a service by combining a plurality of servers having different functions.

  In addition, according to the life and death monitoring system of the present invention, life and death monitoring can be performed in more detail, for example, by specifying a failed server. Further, according to the life and death monitoring method program of the present invention, the computer or computer system can execute the life and death monitoring method of the present invention.

It is a figure showing an example of composition of life and death monitoring system 1 of a 1st embodiment. It is a figure which shows the specific example of the alive monitoring system. It is a figure which shows the operation | movement sequence of the alive monitoring system. It is a figure which shows the example of the response signal of the alive monitoring system. It is a diagram showing an operation sequence when the voice DB server 10 ₂ of life-and-death monitoring system 1 fails. Is a diagram showing an example of a response signal when the voice DB server 10 ₂ fails. Is a diagram showing an operation sequence in a case where the storage 10 ₃ of life-and-death monitoring system 1 fails. It is a figure which shows the operation | movement flow of the alive monitoring apparatus. In life-and-death monitoring system 2 of the second embodiment is a view showing an operation sequence when the voice DB server ₂₁₀₂ has failed. Is a diagram showing an example of a response signal when the voice DB server ₂₁₀₂ has failed. It is a diagram showing an operation sequence when the storage 210 ₃ of life-and-death monitoring system 2 has failed. It is a diagram showing an example of a response signal when the storage 210 ₃ of life-and-death monitoring system 2 has failed. It is a figure which shows the operation | movement flow of the alive monitoring apparatus.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The same reference numerals are given to the same components in a plurality of drawings, and the description will not be repeated.

[First Embodiment]
In FIG. 1, the example of a structure of the alive monitoring system 1 of 1st Embodiment of this invention is shown. The life and death monitoring system 1 of this embodiment includes a multi-server system 100 and a life and death monitoring device 30.

The multi-server system 100 includes a plurality of servers 10 ₁ to 10 _n having different functions. The servers 10 ₁ to 10 _n may include a plurality of servers having the same function. When there are a plurality of servers having the same function, a load balancer 20 is provided for the purpose of distributing processing to each server. The load balancer 20 is not an essential configuration.

  The alive monitoring device 30 is a device that requests alive monitoring of the multi-server system 100 by transmitting a request signal for a service provided by the multi-server system 100 to any of a plurality of servers. The life / death monitoring apparatus 30 includes a life / death monitoring request transmission unit 31 and a life / death monitoring unit 32.

  The life / death monitoring request transmission unit 31 transmits a request signal for a service provided by the multi-server system 100 as a life / death monitoring signal of the multi-server system 100 that provides a service by combining a plurality of servers having different functions. This request signal is the same as a signal for requesting a normal service used by the multi-server system 100. Hereinafter, the request signal is referred to as a service request signal.

  The alive monitoring unit 32 receives a response signal to the service request signal, and uses the status of the destination server of the request signal included in the response signal and the status of the cooperation destination server in cooperation with the destination server. Monitor server aliveness.

  The response signal is a response signal for a processing request to the server. For example, a response signal of a communication protocol (HTTP) used for transmission / reception of content such as HTML between a web browser and a web server is a signal including a header portion and a body portion. The header part and the body part include a status code indicating the state of the server. Details of the response signal and the status signal will be described later.

  According to the alive monitoring device 30 of the alive monitoring system 1, the service request signal normally used by the multi-server system 100 is used as a request signal for executing the alive monitoring. And the normality of each server which comprises the multi-server system 100 is monitored based on the status code of the response signal with respect to a service request signal. Therefore, normality including cooperative operations of a plurality of servers having different functions can be monitored without changing the configuration on the multi-server system 100 side.

  This will be described more specifically. FIG. 2 shows a functional configuration example when the multi-server system 100 of the alive monitoring system 1 of the present embodiment is a safety confirmation system. The safety confirmation system (hereinafter referred to as the multi-server system 100) is a system that provides services when communication of safety confirmation to the disaster area increases due to the occurrence of a disaster such as an earthquake or an eruption, and communication becomes difficult. It is.

  A user who wants to check a message inputs the telephone number of the disaster area in the multi-server system 100, so that the message registered by the disaster victim by voice or text can be acquired by voice. Note that when the multi-server system 100 is used for the purpose of original safety confirmation, a server (media server) (not shown) and a terminal operated by the user are connected to the multi-server system 100.

The multi-server system 100 includes a voice AP server 10 ₁ , a voice DB server 10 ₂ , a storage 10 ₃ , a Web-AP server 10 ₄ , and a voice conversion server 10 ₅ . Each server constituting the multi-server system 100 is connected via a network. In FIG. 2, the notation of the network connecting each server and the network connecting the multi-server system 100 and the alive monitoring device 30 is omitted.

Voice AP server 10 ₁ performs data acquisition request to the voice DB server 10 ₂ If there is a message confirmation request from the user. If registered message is voice perform data acquisition request to the voice DB server 10 _2. Further, if the message that has been registered in the text perform data acquisition request to the Web-AP server 10 _4.

Voice DB server 10 ₂ searches the audio data stored in the storage 10 ₃ to the data acquisition request from the voice AP server 10 _1. Storage 10 ₃ accumulates the message data by voice victims has registered.

Web-AP server 10 ₄ stores the message data by the text that victims have registered. Voice conversion server ₁₀₅ converts the message data by the text-to-speech in accordance with the requirements of the voice AP server 10 _1.

When the message corresponding to the message acquisition request from the user is registered by voice, the multi-server system 100 operates in cooperation with the voice AP server 10 ₁ , voice DB server 10 ₂ , and storage 10 _3. Play the message. If the message is registered as text, the voice AP server 10 ₁ , the Web-AP server 10 ₄ , and the voice conversion server 10 ₅ operate in cooperation to reproduce the message.

Thus, the multi-server system 100 is a system that provides a service by combining a plurality of servers having different functions. Incidentally, if the number of message to be registered is large, a plurality of voice AP server 10 ₁ is provided, it is possible to distribute the load on each of the voice AP server 10 _1. When comprising a plurality of speech AP server 10 _1, the load balancer 20 is provided. Load balancer 20 distributes the message acquisition request from the user to each of the audio AP server 10 _1.

The life and death monitoring request transmission unit 31 according to the present embodiment transmits a service request signal to _one of the plurality of voice AP servers 101 (any _one of the voice AP servers indicated by a broken line in FIG. 2). The service request signal is transmitted to any server that starts processing first to realize the service provided by the multi-server system when alive monitoring of the multi-server system other than the safety confirmation system is performed.

  Hereinafter, the operation of the alive monitoring system 1 will be described by taking as an example a case where a message is registered by voice.

〔Normal〕
The operation of the alive monitoring system 1 when the multi-server system 100 is normal will be described with reference to FIG. Down monitoring device 30 performs alive monitoring request to the voice AP server 10 ₁ (step S1). This life and death monitoring request is made, for example, by inputting a specific fixed telephone number of the disaster area. This fixed telephone number may be any fixed telephone number in the disaster area. In this way, alive monitoring can be performed by using the service request signal of the service provided by the multi-server system 100.

  For the purpose of discriminating between test signals, a test telephone number (for example, 9999999999 may be input. If a test telephone number is set and a test message is registered, Life and death monitoring can be performed without affecting other message information.

Voice AP server 10 ₁ receives the life-and-death monitoring request (service request signal), and transmits the audio data acquisition request to the voice DB server 10 ₂ (step S2). Voice DB server 10 ₂ sends the audio data acquisition request to the storage 10 ₃ receives the audio data acquisition request (step S3).

Voice DB server 10 _2, acquires the audio data from the storage 10 ₃ (step S4), and transmits the voice data to the voice AP server 10 ₁ (step S5). Voice AP server 10 ₁ may store received audio data in the body portion of the response signal and transmits the life-and-death monitoring device 30 (step S6). Response signal transmitted to the life-and-death monitoring device 30 from the sound AP server 10 ₁ in step S6, a response signal to the service request signal in step S1.

  FIG. 4 shows an example of the response signal. The example of FIG. 4 is an HTTP response signal. The response signal is composed of a header part and a body part. A status code is included in the header part and the body part.

The first line of the header portion describes a status code "200" as an HTTP version information as "HTTP / 1.1" indicating the state of the voice AP server 10 _1. "200" of the status code, alive monitoring request in step S1 is meant that it has been successfully processed (OK) voice AP server 10 _1.

The third line of the body part is a status code indicating the status of the voice DB server 10 _2. In this example, “200” is described because it is normal.

Thus, in the header part of the response signal, the status code of the destination server to which the alive monitoring device 30 has transmitted the service request signal is described. Further, the body portion, the voice AP server 10 ₁ is described status code of the voice DB server 10 ₂ to transmit the voice data acquisition request. That is, the status code of the destination server that transmitted the service request signal is described in the header part, and the status code of the cooperation destination server that cooperates with the destination server is described in the body part.

  The alive monitoring device 30 performs alive monitoring processing of the multi-server system 100 based on the status code of the header part and body part of the response signal (step S20).

[In case of failure]
Figure 5 shows an operation sequence when the voice DB server 10 ₂ is faulty. When voice DB server 10 ₂ is faulty, the audio data acquisition request to the storage 10 ₃ from the voice DB server 10 ₂ (step S3) is not transmitted. Of course, there is no response to step S3 (step S4).

Voice DB server 10 ₂ in this case, sends a status code representing the server error in the body portion of the response signal, for example by adding "500" to the voice AP server 10 ₁ (step S7). FIG. 6 shows an example of the response signal in this case. “500” is described in the third line of the body part. “500” is a status code indicating an internal server error.

When voice DB server 10 ₂ is faulty, the response signal shown in FIG. 6, for example from speech AP server 10 ₁ is transmitted to the life-and-death monitoring device 30 (step S8). The alive monitoring device 30 can monitor whether each server constituting the multi-server system 100 is normal or faulty based on the header part of the response signal and the status code of the body part.

  Status codes representing server errors include the 400s and 500s. 4XX is a status code added when there is a cause of error on the side (client) that requested processing. Further, 5XX is a status code added when there is a cause of error on the side (server) requested to process.

Figure 7 shows an operation sequence in a case where the storage 10 ₃ is faulty. In this case, the storage 10 ₃ transmits to the body portion of the response signal status codes representing a server error, for example by adding "500" to the voice DB server 10 ₂ (step S7). The response signal is transmitted as it is to the alive monitoring device 30 (step S8). The response signal in this case is the same as the example shown in FIG. As described above, when _one of the voice DB server 10 ₂ and the storage 10 ₃ linked with the voice AP server 101 is out of order, a status code indicating a server error is added to the body portion of the response signal.

  FIG. 8 shows an operation flow of the life and death monitoring process (step S20) performed by the life and death monitoring unit 32. The life and death monitoring unit 32 performs life and death monitoring using the status code.

  In the life and death monitoring unit 32, the status code of the header part is in the 200s (2XX) (YES in step S21), the status code of the body part is in the 200s (2XX) (YES in step S21), and the multiserver system 100 is normal. If it is determined that there is, the process is terminated (step S25). A status code in the 200s means success. For example, 201 means creation completion (Created), and 202 means acceptance (Accepted).

If the status code of the header part is not 200s, and ends the process determines that the voice AP server 10 ₁ is faulty (Step S22). Status code 200 series header portion, and, if the status code of the body portion is not 200s, audio DP server 10 ₂ fails or ends the process determines that the storage ₁₀₃ is faulty (Step S24).

  As described above, according to the life and death monitoring system 1 of the present embodiment, the normality of the cooperative operation between servers of the multi-server system 100 that provides a service by combining a plurality of servers having different functions is monitored. be able to.

[Second Embodiment]
Next, the life and death monitoring system 2 according to the second embodiment of the present invention in which the failure location can be specified will be described. The life and death monitoring system 2 according to the present embodiment is different from the life and death monitoring system 1 in that each server constituting the multi-server system 100 adds failure information to a response signal to a service request signal.

Life-and-death monitoring system 2 is comprised of a life-and-death monitoring device 230, a voice AP server 210 _1, a voice DB server _2102, a storage _2103. Configure a voice AP server 210 _1, a voice DB server _2102, a multi-server system 200 in the storage 210 _3.

The voice AP server 210 ₁ , the voice DB server 210 _2, and the storage 210 ₃ are different from the servers described above (FIG. 3) in that they include failure information addition units 211 ₁ , 211 ₂ , and 211 ₃ , respectively. The failure information addition units 211 ₁ , 211 ₂ , 211 ₃ add failure information to the body part of the response signal to the service request signal. Since the functional configuration of the life and death monitoring system 2 is the same as the functional configuration of the life and death monitoring system 1, the illustration thereof is omitted.

9 shows an operation sequence of the life-and-death monitoring system 2 when the voice DB server ₂₁₀₂ has failed. Operation sequence when the voice DB server ₂₁₀₂ has failed, the same as the above example (FIG. 5), the audio data acquisition request to the storage 210 ₃ from the voice DB server ₂₁₀₂ (step S3) is not executed. Of course, there is no response to step S3 (step S4).

In this case, the failure information addition unit 211 ₂ of the voice DB server ₂₁₀₂ adds the failure information in the body portion of the response signal to the service request signal (step S1) (step S11). Response signal failure information is added is transmitted to the voice AP server 210 ₁ (step S12).

FIG. 10 shows an example of a response signal to which failure information is added. On the eleventh line of the body part, a failure server ID “voice-db-xxx” for identifying the failed server is added after timestamp. The response signal as a response signal of a service request signal from the voice AP server 210 ₁ (step S1), the sent to the life-and-death monitoring device 230 (step S13).

  In this way, the failure server ID is included as failure information in the response signal. The failure information may include information indicating the failure content in addition to the identifier for identifying the server. For example, it is possible to monitor the state of the multi-server system 100 in more detail by adding failure information such as “memory shortage”.

Further, the malfunction information of the voice DB server ₂₁₀₂ may be added by the speech AP server 210 _1. For example, due to failure of the network that connects the speech AP server 210 ₁ and the voice DB server _2102, there are cases where the response signal from the voice DB server ₂₁₀₂ can not be obtained. In that case, the timeout is detected in the voice AP server 210 _1, a failure information adding unit 211 ₁ of the voice AP server 210 _1, may be added to the failure information of the voice DB server ₂₁₀₂ (step S14).

11 shows an operation sequence of the life-and-death monitoring system 2 when the storage 210 ₃ fails. In this case, the storage 210 _3, additional failure information in the body portion of the response signal for the speech data acquisition request (step S15) and sends to the voice DB server ₂₁₀₂ (step S16).

FIG. 12 shows an example of a response signal including the failure information added in step S15. The failure server ID “strage-xxx” is added to the 11th line of the body part. The response signal, via the voice DB server ₂₁₀₂ and the speech AP server 210 _1, are sent to the life-and-death monitoring device 230 (step S17, S18). Failure information in the storage 210 _3, as well as the failure information of the voice DB server ₂₁₀₂ may be added by the error information adding unit 211 and _second sound DB server ₂₁₀₂ (step S18). The alive monitoring device 230 that has received the response signal to which the failure information is added can perform the alive monitoring process (step S30) in more detail.

  The life and death monitoring process (step S30) will be described with reference to FIG. Processing when the status code of the header portion of the response signal is in the 200s (YES in step S21) and the status code of the body part is not in the 200s (NO in step S23) is different from the above operation flow (FIG. 8).

The alive monitoring device 230 refers to the failure information of the body part of the response signal (step S31). If failed server ID representing the failure information is "for voice-db", life-and-death monitoring device 230 ends the process determines that the voice DB server ₂₁₀₂ has failed (step S32). If failed server ID is "strage", life-and-death monitoring device 230 ends the process determines that the storage 210 ₃ has failed (step S33).

  As described above, according to the life and death monitoring system 2 of the present embodiment, the failure location of the multi-server system 100 can be specified. In addition, this invention is not limited to said embodiment. The multi-server system 100 of the present embodiment has been described by taking the safety confirmation system as an example, but may be another multi-server system as long as it provides a service by combining servers having different functions.

  Although the HTTP response signal has been described as an example, the present invention can also be applied to communication based on other protocols such as SIP (Session Initiation Protocol). As described above, the present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the gist.

  When the above device (server) is realized by a computer, the processing contents of functions that each device should have are described by a program. Then, by executing this program on the computer, the processing unit in each device is realized on the computer.

  The program is distributed by selling, transferring, or lending a portable recording medium such as a DVD or CD-ROM in which the program is recorded. Further, the program may be distributed by storing the program in a recording device of the server computer and transferring the program from the server computer to another computer via a network.

  Each means may be configured by executing a predetermined program on a computer, or at least a part of these processing contents may be realized by hardware.

1: Life and death monitoring system 10 ₁ to 10 _n : Server 20: Load balancer 30: Life and death monitoring device 31: Life and death monitoring request transmission unit 32: Life and death monitoring unit 100: Multi-server system

Claims (8)

A life / death monitoring request for transmitting a service request signal provided by the multi-server system to any of the plurality of servers as a life / death monitoring signal for a multi-server system that provides a service by combining a plurality of servers having different functions. A transmission unit;
A response signal to the request signal is received, and the status of the request server included in the response signal and the status of the cooperation destination server that cooperates with the destination server are used to monitor the aliveness of the plurality of servers A life / death monitoring apparatus comprising: A life and death monitoring system comprising a multi-server system that provides a service by combining a plurality of servers having different functions, and a life and death monitoring device,
The alive monitoring device is:
As a life and death monitoring signal of the multi-server system, a life and death monitoring request transmission unit that transmits a service request signal provided by the multi-server system to any of the plurality of servers,
A plurality of servers that receive a response signal to the request signal and use the status of the destination server of the request signal included in the response signal, the status of the cooperation destination server that cooperates with the destination server, and failure information A life and death monitoring unit that performs life and death monitoring of
The server
A life and death monitoring system comprising: a failure information adding unit that adds the failure information to the response signal to the request signal of the service. In the life and death monitoring system according to claim 2,
The life and death monitoring system, wherein the failure information includes an identifier for identifying the cooperation destination server. A life and death monitoring method performed by a life and death monitoring device,
A life / death monitoring request for transmitting a service request signal provided by the multi-server system to any of the plurality of servers as a life / death monitoring signal for a multi-server system that provides a service by combining a plurality of servers having different functions. Sending step;
A response signal to the request signal is received, and the status of the request server included in the response signal and the status of the cooperation destination server that cooperates with the destination server are used to monitor the aliveness of the plurality of servers A life and death monitoring method comprising: performing a life and death monitoring step. A life / death monitoring method performed by a life / death monitoring system including a multi-server system that provides a service by combining a plurality of servers having different functions, and a life / death monitoring apparatus,
The life and death monitoring device performs a life and death monitoring request transmission step of transmitting the service request signal as a signal for life and death monitoring of the multi-server system,
The server performs a failure information addition step of adding failure information to a response signal to the request signal,
The alive monitoring device receives a response signal to the request signal, and uses the status of the transmission destination server of the request signal included in the response signal, the status of the cooperation destination server linked with the transmission destination server, and failure information A life / death monitoring step for performing life / death monitoring of a plurality of the servers. In the life and death monitoring method according to claim 5,
The life and death monitoring method, wherein the failure information includes an identifier for identifying the cooperation destination server.   A life and death monitoring method program for causing a computer to execute the life and death monitoring method according to claim 4.   A life / death monitoring method program for causing a computer system to execute the life / death monitoring method according to claim 5.

Images