WO2015170365A1 - Server device and computer system - Google Patents

Server device and computer system Download PDF

Info

Publication number
WO2015170365A1
WO2015170365A1 PCT/JP2014/002435 JP2014002435W WO2015170365A1 WO 2015170365 A1 WO2015170365 A1 WO 2015170365A1 JP 2014002435 W JP2014002435 W JP 2014002435W WO 2015170365 A1 WO2015170365 A1 WO 2015170365A1
Authority
WO
WIPO (PCT)
Prior art keywords
server
application
slave
completed
watchdog timer
Prior art date
Application number
PCT/JP2014/002435
Other languages
French (fr)
Japanese (ja)
Inventor
誠司 関
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to PCT/JP2014/002435 priority Critical patent/WO2015170365A1/en
Priority to TW103116947A priority patent/TW201543224A/en
Publication of WO2015170365A1 publication Critical patent/WO2015170365A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units

Definitions

  • the present invention relates to a computer system composed of a master server and a slave server, and relates to a technique for initializing a slave server when a communication error occurs between the master server and the slave server.
  • an initialization command is transmitted from the master device to the slave device, the watchdog timer is initialized and counted down every time the slave device receives the initialization command, and an initialization command (
  • the following patent document discloses a control method for initializing (resetting) a slave device when the watchdog timer times out without receiving a clear command.
  • the computer system control method disclosed in the above patent document assumes a case where a master device and a slave device communicate one-to-one, and in a network in which a plurality of slave devices executing a plurality of applications exist, the slave device
  • this is not intended for a system that performs data communication with other slave devices, a situation occurs in which the slave device is initialized regardless of the data reception status between the slave devices.
  • FIG. 5 is a diagram for explaining a problem in resetting a slave device in a computer system having a plurality of slave devices.
  • the slave device 1 that has received the input data 1 from the master device executes an application for the input data 1.
  • the next input data 2 cannot be received.
  • the watchdog timer (WDT) of the slave device 1 times out.
  • the slave device 1 when the slave device 1 is reset regardless of the operation status of the application of the slave device 1 and the data transfer status between the slave devices, the output data of the application process for the input data 1 executed in the slave device 1 is displayed. There is a problem in that it is not possible to take over and it is necessary to re-execute an application process that has already been executed.
  • the present invention has been made to solve the above-described problems, and in the case where the slave device is reset when communication between the slave device and the master device is interrupted, the application processing executed in the slave device is performed.
  • An object of the present invention is to provide a computer system capable of taking over results.
  • a computer system is a computer system in which a master server and a slave server that executes an application according to input data transmitted from the master server are connected via a network.
  • the master server periodically transmits a clear command for detecting a communication abnormality with the slave server to the slave server,
  • the slave server initializes the watchdog timer in response to the clear command and starts a countdown. When the watchdog timer times out, the slave server completes all processes of the application that has started executing before the timeout occurs.
  • the other server that has received the output data transmits the received output data to the master server based on the identification flag added to the output data.
  • the processing of the application being executed in the slave server is completed, and the processing is completed after the watchdog timer times out. Since an identification flag indicating completion is added and transmitted to another server device, the processing result of the application executed in the slave device can be taken over when a communication error with the master server occurs.
  • FIG. 1 is a configuration diagram of a computer system according to an embodiment of the present invention.
  • the computer system shown in FIG. 1 is configured by connecting a plurality of slave servers 110 and 120 to a master server 100 via a network.
  • the slave server 120 is assumed to have the same configuration as the slave server 110.
  • the slave server 110 can execute a plurality of applications, and communicates with the master server 100 and the slave server 120.
  • the master server 100 includes a clear command transmission unit 101 that periodically transmits a clear command for detecting a communication abnormality with the slave server 110 and a communication unit 102 that transmits the clear command to the slave server 110.
  • the slave server 110 includes an application monitoring unit 111, a reset unit 112, a watchdog timer (hereinafter referred to as WDT) 113, and a communication unit 114.
  • the slave server 110 can process a plurality of applications, receives input data from the master server 100 for each application, and transmits a processing result of each application as output data to the master server.
  • the communication unit 114 of the slave server 110 transmits and receives data to and from the master server 100 and other slave servers 120.
  • WDT 113 initializes a timer in response to a clear command from master server 100 input via communication unit 114 and starts a countdown. If the clear command is not received within a predetermined period after the countdown is started and time-out occurs, the WDT 113 notifies the application monitoring unit 111 that time-out has occurred.
  • the application monitoring unit 111 monitors the processing status of the application and receives a notification indicating that the WDT 113 has timed out
  • the application monitoring unit 111 confirms that the processing of all the active applications has been completed, and notifies the reset unit 112 accordingly. To do.
  • the reset unit 112 receives the notification from the application monitoring unit 111 and resets the slave server 110.
  • the application monitoring unit 111 transmits output data to the other server 120 via the communication unit 114, to which the identification flag indicating that the processing is completed after the timeout is added to the processing result of the application.
  • the application monitoring unit 111 transmits a notification signal indicating that the processing of all applications has been completed to the master server 100 and the slave server 120 via the communication unit 114.
  • FIG. 3 is a diagram illustrating an example of a format of data output from the slave server 110. As shown in FIG. 3, output data is given an output data ID, a data size, an identification flag, and a sequence number.
  • FIG. 2 is a flowchart showing the operation of the slave server 110 shown in FIG. 1.
  • the slave server 110 is initialized when it hangs up. The operation is shown.
  • Slave server 110 detects whether or not a clear command periodically transmitted from master server 100 has been received (ST1). When the clear command is received, the slave server 110 initializes the WDT 113 and starts counting down (ST2). If the clear command cannot be received within the predetermined period, WDT 113 times out (ST3).
  • the application monitoring unit 111 waits for completion of the application being processed (ST4).
  • the processing of the application is completed (ST5), output data in which an identification flag indicating that the processing result is the processing result after the time-out is added to other predetermined servers (master server 100, slave server 120). (ST6).
  • the slave server 120 that has received the output data with the identification flag added detects the identification flag in the application monitoring unit, and transmits the output data to the master server 100.
  • the application monitoring unit 111 When all the applications being processed in the slave server 110 are terminated (ST7), the application monitoring unit 111 notifies the other server of the completion of the processing of the application via the communication unit 114 (ST8). Further, the application monitoring unit 111 notifies the reset unit 112 that the processing of the application has been completed, and the reset unit 112 resets the slave server 110 (ST9).
  • FIG. 4 is a diagram showing the operation of the computer system shown in FIG. 1, and initializes the slave server 110 when the slave server 110 cannot receive input data from the master server 100 due to a communication error or the like and hangs up. The operation is shown.
  • Master server 100 transmits input data 1 to slave server 110.
  • the slave server 110 that has received the input data 1 performs an application process on the input data 1. If a reception abnormality occurs in the slave server 110 that cannot receive data from the master server 100 during application processing, the next input data 2 cannot be received. Furthermore, since the clear command from the master server 100 cannot be received within a predetermined time, a timeout of the WDT 113 occurs.
  • the application monitoring unit 111 of the slave server 110 completes the application process for the input data 1
  • the output data 1 with the identification flag indicating the processing result after the timeout is added to the processing result as the master server 100, And to the slave server 120.
  • the slave server 120 that has received the output data 1 detects the identification flag and transmits the output data 1 to the master server.
  • the application monitoring unit 111 of the slave server 110 notifies the other server of the completion of application processing via the communication unit 114.
  • the slave server 120 that has received the end notification transmits the received end notification to the master server 100.
  • the reset unit 112 of the slave server 110 resets the slave server 110 in response to the notification from the application monitoring unit 111.
  • the master server 100 has caused the output data 1 for the input data 1 to be abnormal. Can be received via no slave server 120.
  • the slave server 110 can be taken over, it is possible to prevent data inconsistency, and it is not necessary to perform the processing completed on the slave server again. It is possible to suppress a decrease in system performance associated with.

Abstract

The present invention provides a computer system in which the results of processing performed by an application running on a slave device can be taken over when the slave device is reset after the communication between the slave device and the master device has been interrupted. A slave server constituting the computer system of the present invention is provided with an application monitoring unit which, after a watchdog timer has timed out, monitors whether processing initiated by an application prior to the time-out is completed, wherein if the processing is completed, the application monitoring unit outputs a notification signal indicating the completion, and transmits output data to another server device via a network, said output data including both the results of the completed processing performed by the application and an added identification flag which indicates that the processing has been completed after the time-out of the watchdog timer.

Description

サーバ装置、およびコンピュータシステムServer device and computer system
　本発明は、マスタサーバとスレーブサーバとから構成されるコンピュータシステムに関するものであり、マスタサーバとスレーブサーバとの間に通信異常が発生した場合にスレーブサーバを初期化する技術に関する。 The present invention relates to a computer system composed of a master server and a slave server, and relates to a technique for initializing a slave server when a communication error occurs between the master server and the slave server.
　マスタサーバとスレーブサーバとから構成されるコンピュータシステムにおいて、マスタサーバから送信されるデータをスレーブサーバが受信できない通信異常が発生すると、スレーブサーバがハングアップし、マスタサーバがネットワークを介してスレーブサーバを制御することができず、スレーブサーバのリセットができないという問題があった。 In a computer system consisting of a master server and a slave server, if a communication error occurs where the slave server cannot receive data sent from the master server, the slave server hangs up and the master server connects the slave server via the network. There was a problem that it could not be controlled and the slave server could not be reset.
　こうした課題を解決するため、マスタ装置からスレーブ装置に初期化指令を送信し、スレーブ装置において初期化指令を受信する度にウォッチドックタイマの初期化とカウントダウンを行い、通信異常等により初期化指令（クリアコマンド）が受信されずにウォッチドックタイマがタイムアウトしたときにスレーブ装置を初期化（リセット）する制御方法が下記特許文献に開示されている。 In order to solve these problems, an initialization command is transmitted from the master device to the slave device, the watchdog timer is initialized and counted down every time the slave device receives the initialization command, and an initialization command ( The following patent document discloses a control method for initializing (resetting) a slave device when the watchdog timer times out without receiving a clear command.
特開２００６－１１０１５０号公報JP 2006-110150 A 特開２００４－２７４２７５号公報JP 2004-274275 A
　上記特許文献に開示されたコンピュータシステムの制御方法は、マスタ装置とスレーブ装置が一対一で通信する場合を想定しており、複数のアプリケーションを実行する複数のスレーブ装置が存在するネットワークにおいて、スレーブ装置が他のスレーブ装置とデータ通信を行うシステムを想定したものではないため、スレーブ装置間でのデータ受信状況に関係なくスレーブ装置を初期化する事態が生じる。 The computer system control method disclosed in the above patent document assumes a case where a master device and a slave device communicate one-to-one, and in a network in which a plurality of slave devices executing a plurality of applications exist, the slave device However, since this is not intended for a system that performs data communication with other slave devices, a situation occurs in which the slave device is initialized regardless of the data reception status between the slave devices.
　図５は複数のスレーブ装置が存在するコンピュータシステムにおいてスレーブ装置をリセットする際の問題点を説明するための図である。図５に示すように、マスタ装置からの入力データ１を受信したスレーブ装置１は、入力データ１に対するアプリケーションを実行する。アプリケーション処理の実行中、スレーブ装置１においてマスタ装置からのデータを受信できない受信異常が発生すると、次の入力データ２を受信することができない。さらに、マスタ装置からのクリアコマンドを所定時間内に受信することができないため、スレーブ装置１のウォッチドックタイマ（ＷＤＴ）がタイムアウトする。 FIG. 5 is a diagram for explaining a problem in resetting a slave device in a computer system having a plurality of slave devices. As shown in FIG. 5, the slave device 1 that has received the input data 1 from the master device executes an application for the input data 1. During the execution of application processing, if a reception error occurs in the slave device 1 that prevents receiving data from the master device, the next input data 2 cannot be received. Furthermore, since the clear command from the master device cannot be received within a predetermined time, the watchdog timer (WDT) of the slave device 1 times out.
　ここで、スレーブ装置１のアプリケーションの動作状況やスレーブ装置間でのデータの引き継ぎ状況に関係なくスレーブ装置１をリセットした場合、スレーブ装置１で実行していた入力データ１に対するアプリケーション処理の出力データを引き継ぐことができず、既に実行していたアプリケーション処理を再度実行することが必要になるという問題があった。 Here, when the slave device 1 is reset regardless of the operation status of the application of the slave device 1 and the data transfer status between the slave devices, the output data of the application process for the input data 1 executed in the slave device 1 is displayed. There is a problem in that it is not possible to take over and it is necessary to re-execute an application process that has already been executed.
　この発明は上記のような問題点を解決するためになされたもので、スレーブ装置とマスタ装置との通信が途切れた際にスレーブ装置をリセットする場合において、スレーブ装置において実行していたアプリケーションの処理結果を引き継ぐことが可能なコンピュータシステムを提供することを目的とする。 The present invention has been made to solve the above-described problems, and in the case where the slave device is reset when communication between the slave device and the master device is interrupted, the application processing executed in the slave device is performed. An object of the present invention is to provide a computer system capable of taking over results.
　本発明に係わるコンピュータシステムは、マスタサーバと、当該マスタサーバから送信される入力データに応じてアプリケーションを実行するスレーブサーバとがネットワークを介して接続されたコンピュータシステムであって、
　前記マスタサーバは、前記スレーブサーバとの通信異常を検出するためのクリアコマンドを定期的に前記スレーブサーバに送信し、
　前記スレーブサーバは、前記クリアコマンドに応じてウォッチドックタイマを初期化するとともにカウントダウンを開始し、当該ウォッチドックタイマがタイムアウトした場合、タイムアウト発生までに実行を開始しているアプリケーションの全ての処理を完了し、完了したアプリケーションの処理結果に当該処理が前記ウォッチドックタイマのタイムアウト後に完了したことを示す識別フラグを付加した出力データを、前記ネットワークを介して他のサーバに送信し、
　前記出力データを受信した他のサーバは、当該出力データに付加された前記識別フラグに基づき受信した出力データを前記マスタサーバに送信する。 A computer system according to the present invention is a computer system in which a master server and a slave server that executes an application according to input data transmitted from the master server are connected via a network.
The master server periodically transmits a clear command for detecting a communication abnormality with the slave server to the slave server,
The slave server initializes the watchdog timer in response to the clear command and starts a countdown. When the watchdog timer times out, the slave server completes all processes of the application that has started executing before the timeout occurs. And sending output data with an identification flag indicating that the processing is completed after the watchdog timer has timed out to the processing result of the completed application to another server via the network,
The other server that has received the output data transmits the received output data to the master server based on the identification flag added to the output data.
　本発明に係わるコンピュータシステムによれば、スレーブサーバのウォッチドックタイマがタイムアウトした場合、スレーブサーバにおいて実行中のアプリケーションの処理を完了し、完了したアプリケーションの処理結果に当該処理がウォッチドックタイマのタイムアウト後に完了したことを示す識別フラグを付加して他のサーバ装置に送信するので、マスタサーバとの通信異常が発生した際、スレーブ装置において実行していたアプリケーションの処理結果を引き継ぐことができる。 According to the computer system according to the present invention, when the watchdog timer of the slave server times out, the processing of the application being executed in the slave server is completed, and the processing is completed after the watchdog timer times out. Since an identification flag indicating completion is added and transmitted to another server device, the processing result of the application executed in the slave device can be taken over when a communication error with the master server occurs.
本発明の一実施形態に係わるコンピュータシステムの構成図である。It is a block diagram of the computer system concerning one Embodiment of this invention. スレーブサーバの動作を示すフロー図である。It is a flowchart which shows operation | movement of a slave server. スレーブサーバの出力データフォーマットの一例を示す図である。It is a figure which shows an example of the output data format of a slave server. 本発明の一実施形態に係わるコンピュータシステムの動作を示す図である。It is a figure which shows operation | movement of the computer system concerning one Embodiment of this invention. コンピュータシステムの動作を示す図である。It is a figure which shows operation | movement of a computer system.
　図１は本発明の一実施形態に係わるコンピュータシステムの構成図である。図１に示すコンピュータシステムは、マスタサーバ１００にネットワークを介して複数のスレーブサーバ１１０、１２０を接続して構成される。スレーブサーバ１２０はスレーブサーバ１１０と同様の構成を有するものとする。 FIG. 1 is a configuration diagram of a computer system according to an embodiment of the present invention. The computer system shown in FIG. 1 is configured by connecting a plurality of slave servers 110 and 120 to a master server 100 via a network. The slave server 120 is assumed to have the same configuration as the slave server 110.
　スレーブサーバ１１０は複数のアプリケーションを実行可能であり、マスタサーバ１００、およびスレーブサーバ１２０との間で通信を行う。マスタサーバ１００は、スレーブサーバ１１０との通信異常を検出するためのクリアコマンドを定期的に送信するクリアコマンド送信部１０１、クリアコマンドをスレーブサーバ１１０に送信する通信部１０２を備える。 The slave server 110 can execute a plurality of applications, and communicates with the master server 100 and the slave server 120. The master server 100 includes a clear command transmission unit 101 that periodically transmits a clear command for detecting a communication abnormality with the slave server 110 and a communication unit 102 that transmits the clear command to the slave server 110.
　スレーブサーバ１１０は、アプリ監視部１１１、リセット部１１２、ウォッチドックタイマ（以下、ＷＤＴ）１１３、通信部１１４を備える。スレーブサーバ１１０は複数のアプリケーションを処理することが可能であり、マスタサーバ１００からの入力データをアプリケーション毎に受信し、各アプリケーションの処理結果を出力データとしてマスタサーバへ送信する。 The slave server 110 includes an application monitoring unit 111, a reset unit 112, a watchdog timer (hereinafter referred to as WDT) 113, and a communication unit 114. The slave server 110 can process a plurality of applications, receives input data from the master server 100 for each application, and transmits a processing result of each application as output data to the master server.
　スレーブサーバ１１０の通信部１１４は、マスタサーバ１００、および他のスレーブサーバ１２０との間でデータを送受信する。ＷＤＴ１１３は、通信部１１４を介して入力されるマスタサーバ１００からのクリアコマンドに応じてタイマを初期化してカウントダウンを開始する。カウントダウン開始後、所定期間内にクリアコマンドが受信できずにタイムアウトした場合、ＷＤＴ１１３はアプリ監視部１１１にタイムアウトしたことを通知する。 The communication unit 114 of the slave server 110 transmits and receives data to and from the master server 100 and other slave servers 120. WDT 113 initializes a timer in response to a clear command from master server 100 input via communication unit 114 and starts a countdown. If the clear command is not received within a predetermined period after the countdown is started and time-out occurs, the WDT 113 notifies the application monitoring unit 111 that time-out has occurred.
　アプリ監視部１１１はアプリケーションの処理状況を監視し、ＷＤＴ１１３がタイムアウトしたことを示す通知を受信した場合、動作中の全てのアプリケーションの処理が完了したことを確認し、その旨をリセット部１１２に通知する。リセット部１１２は、アプリ監視部１１１からの通知を受けてスレーブサーバ１１０をリセットする。また、アプリ監視部１１１はアプリケーションの処理結果に当該処理がタイムアウト後に完了したものであることを示す識別フラグを付加した出力データを、通信部１１４を介して他のサーバ１２０に送信する。さらにアプリ監視部１１１は、全てのアプリケーションの処理が終了したことを示す通知信号を、通信部１１４を介してマスタサーバ１００、スレーブサーバ１２０に送信する。 When the application monitoring unit 111 monitors the processing status of the application and receives a notification indicating that the WDT 113 has timed out, the application monitoring unit 111 confirms that the processing of all the active applications has been completed, and notifies the reset unit 112 accordingly. To do. The reset unit 112 receives the notification from the application monitoring unit 111 and resets the slave server 110. In addition, the application monitoring unit 111 transmits output data to the other server 120 via the communication unit 114, to which the identification flag indicating that the processing is completed after the timeout is added to the processing result of the application. Furthermore, the application monitoring unit 111 transmits a notification signal indicating that the processing of all applications has been completed to the master server 100 and the slave server 120 via the communication unit 114.
　図３はスレーブサーバ１１０から出力されるデータのフォーマットの一例を示す図である。図３に示すように、出力データには、出力データＩＤ、データサイズ、識別フラグ、シーケンス番号が付与される。 FIG. 3 is a diagram illustrating an example of a format of data output from the slave server 110. As shown in FIG. 3, output data is given an output data ID, a data size, an identification flag, and a sequence number.
　図２は、図１に示すスレーブサーバ１１０の動作を示すフロー図であり、通信異常等によりマスタサーバ１００からの入力データを受信できず、ハングアップした場合にスレーブサーバ１１０を初期化する際の動作を示す。 FIG. 2 is a flowchart showing the operation of the slave server 110 shown in FIG. 1. When the input data from the master server 100 cannot be received due to a communication error or the like and the slave server 110 is initialized when it hangs up. The operation is shown.
　スレーブサーバ１１０はマスタサーバ１００から定期的に送信されるクリアコマンドが受信されたか否かを検知する（ＳＴ１）。クリアコマンドが受信された場合、スレーブサーバ１１０はＷＤＴ１１３を初期化してカウントダウンを開始する（ＳＴ２）。クリアコマンドが所定期間内に受信できない場合、ＷＤＴ１１３がタイムアウトする（ＳＴ３）。 Slave server 110 detects whether or not a clear command periodically transmitted from master server 100 has been received (ST1). When the clear command is received, the slave server 110 initializes the WDT 113 and starts counting down (ST2). If the clear command cannot be received within the predetermined period, WDT 113 times out (ST3).
　ＷＤＴ１１３がタイムアウトした場合、アプリ監視部１１１は処理中のアプリケーションの完了を待つ（ＳＴ４）。アプリケーションの処理が完了すると（ＳＴ５）、処理結果にタイムアウト後の処理結果であることを示す識別フラグを付加した出力データを、予め定められた他のサーバ（マスタサーバ１００、スレーブサーバ１２０）に送信する（ＳＴ６）。ここで、識別フラグが付加された出力データを受信したスレーブサーバ１２０はアプリ監視部にて識別フラグを検出し、出力データをマスタサーバ１００に送信する。 If the WDT 113 times out, the application monitoring unit 111 waits for completion of the application being processed (ST4). When the processing of the application is completed (ST5), output data in which an identification flag indicating that the processing result is the processing result after the time-out is added to other predetermined servers (master server 100, slave server 120). (ST6). Here, the slave server 120 that has received the output data with the identification flag added detects the identification flag in the application monitoring unit, and transmits the output data to the master server 100.
　スレーブサーバ１１０において処理中の全てのアプリケーションが終了した場合（ＳＴ７）、アプリ監視部１１１は、通信部１１４を介して他のサーバにアプリケーションの処理完了を通知する（ＳＴ８）。また、アプリ監視部１１１は、アプリケーションの処理が完了したことをリセット部１１２に通知し、リセット部１１２はスレーブサーバ１１０をリセットする（ＳＴ９）。 When all the applications being processed in the slave server 110 are terminated (ST7), the application monitoring unit 111 notifies the other server of the completion of the processing of the application via the communication unit 114 (ST8). Further, the application monitoring unit 111 notifies the reset unit 112 that the processing of the application has been completed, and the reset unit 112 resets the slave server 110 (ST9).
　図４は、図１に示すコンピュータシステムの動作を示す図であり、通信異常等によりスレーブサーバ１１０がマスタサーバ１００からの入力データを受信できず、ハングアップした場合にスレーブサーバ１１０を初期化する際の動作を示す。 FIG. 4 is a diagram showing the operation of the computer system shown in FIG. 1, and initializes the slave server 110 when the slave server 110 cannot receive input data from the master server 100 due to a communication error or the like and hangs up. The operation is shown.
　マスタサーバ１００は入力データ１をスレーブサーバ１１０に送信する。入力データ１を受信したスレーブサーバ１１０は、入力データ１に対するアプリケーションの処理を行う。アプリケーションの処理を実施中、スレーブサーバ１１０において、マスタサーバ１００からのデータを受信できない受信異常が発生すると、次の入力データ２を受信することができない。さらに、マスタサーバ１００からのクリアコマンドを所定時間内に受信することができないため、ＷＤＴ１１３のタイムアウトが発生する。 Master server 100 transmits input data 1 to slave server 110. The slave server 110 that has received the input data 1 performs an application process on the input data 1. If a reception abnormality occurs in the slave server 110 that cannot receive data from the master server 100 during application processing, the next input data 2 cannot be received. Furthermore, since the clear command from the master server 100 cannot be received within a predetermined time, a timeout of the WDT 113 occurs.
　ＷＤＴ１１３がタイムアウトすると、スレーブサーバ１１０のアプリ監視部１１１は入力データ１に対するアプリケーション処理が完了すると、処理結果にタイムアウト後の処理結果であることを示す識別フラグを付加した出力データ１をマスタサーバ１００、およびスレーブサーバ１２０に送信する。出力データ１を受信したスレーブサーバ１２０は、識別フラグを検出し、出力データ１をマスタサーバへ送信する。 When the WDT 113 times out, when the application monitoring unit 111 of the slave server 110 completes the application process for the input data 1, the output data 1 with the identification flag indicating the processing result after the timeout is added to the processing result as the master server 100, And to the slave server 120. The slave server 120 that has received the output data 1 detects the identification flag and transmits the output data 1 to the master server.
　スレーブサーバ１１０のアプリケーション処理が完了すると、スレーブサーバ１１０のアプリ監視部１１１は、通信部１１４を介して他のサーバにアプリケーションの処理完了を通知する。終了通知を受信したスレーブサーバ１２０はマスタサーバ１００に受信した終了通知を送信する。そして、スレーブサーバ１１０のリセット部１１２はアプリ監視部１１１の通知に応じてスレーブサーバ１１０をリセットする。 When the application processing of the slave server 110 is completed, the application monitoring unit 111 of the slave server 110 notifies the other server of the completion of application processing via the communication unit 114. The slave server 120 that has received the end notification transmits the received end notification to the master server 100. Then, the reset unit 112 of the slave server 110 resets the slave server 110 in response to the notification from the application monitoring unit 111.
　以上のように、スレーブサーバ１１０をリセットする前にアプリケーション処理を完了し、処理結果を他のスレーブサーバに送信することにより、マスタサーバ１００は、入力データ１に対する出力データ１を異常の発生していないスレーブサーバ１２０を経由して受信することができる。 As described above, by completing the application processing before resetting the slave server 110 and transmitting the processing result to another slave server, the master server 100 has caused the output data 1 for the input data 1 to be abnormal. Can be received via no slave server 120.
　これにより、スレーブサーバ１１０で処理していたデータを引き継ぐことができるので、データの不整合を防ぐとともに、スレーブサーバで完了していた処理を再度行う必要がなくなるので、通信異常発生時の初期化に伴うシステム性能の低下を抑制することができる。 As a result, since the data processed by the slave server 110 can be taken over, it is possible to prevent data inconsistency, and it is not necessary to perform the processing completed on the slave server again. It is possible to suppress a decrease in system performance associated with.
１００　マスタサーバ、１０１　クリアコマンド送信部、１１０，１２０，スレーブサーバ、１１１　アプリ監視部、１１２　リセット部、１１３　ＷＤＴ、　１０２，１１４　通信部 100 master server, 101 clear command transmission unit, 110, 120, slave server, 111 application monitoring unit, 112 reset unit, 113 WDT, 102, 114 communication unit

Claims (5)

  1. 　マスタサーバから送信される入力データに応じてアプリケーションを実行するサーバ装置であって、
    　前記マスタサーバから定期的に送信されるクリアコマンドを受信し、当該クリアコマンドに応じて初期化され、カウントダウンを開始するウォッチドックタイマと、
    　前記ウォッチドックタイマがタイムアウトした場合、タイムアウト発生までに実行を開始しているアプリケーションの処理が完了したか否かを監視し、当該処理が完了した際にその旨を示す通知信号を出力するとともに、完了したアプリケーションの処理結果に、当該処理が前記ウォッチドックタイマのタイムアウト後に完了したことを示す識別フラグを付加した出力データを、ネットワークを介して他のサーバ装置に送信するためのアプリケーション監視部と、を備えたサーバ装置。     A server device that executes an application according to input data transmitted from a master server,
    A watchdog timer that receives a clear command periodically transmitted from the master server, is initialized according to the clear command, and starts a countdown;
    When the watchdog timer times out, it monitors whether or not the processing of the application that has started execution before the time-out occurs, and outputs a notification signal indicating that when the processing is completed, An application monitoring unit for transmitting output data to which the identification flag indicating that the process is completed after the watchdog timer time-out is added to the processing result of the completed application to another server device via the network; A server device comprising:
  2. 　前記アプリケーション監視部から出力される通知信号に応じてリセット処理を行うリセット部をさらに備えることを特徴とする請求項１に記載のサーバ装置。 The server device according to claim 1, further comprising a reset unit that performs a reset process in response to a notification signal output from the application monitoring unit.
  3. 　前記アプリケーション監視部は、前記ウォッチドックタイマのタイムアウト後にアプリケーション処理が完了したことを示す通知信号を、ネットワークを介して他のサーバに送信することを特徴とする請求項１に記載のサーバ装置。 The server apparatus according to claim 1, wherein the application monitoring unit transmits a notification signal indicating that the application process is completed after the watchdog timer has timed out to another server via the network.
  4. 　マスタサーバと、当該マスタサーバから送信される入力データに応じてアプリケーションを実行するスレーブサーバとがネットワークを介して接続されたコンピュータシステムであって、
    　前記マスタサーバは、前記スレーブサーバとの通信異常を検出するためのクリアコマンドを定期的に前記スレーブサーバに送信し、
    　前記スレーブサーバは、前記クリアコマンドに応じてウォッチドックタイマを初期化するとともにカウントダウンを開始し、当該ウォッチドックタイマがタイムアウトした場合、タイムアウト発生までに実行を開始しているアプリケーションの処理を完了し、完了したアプリケーションの処理結果に当該処理が前記ウォッチドックタイマのタイムアウト後に完了したことを示す識別フラグを付加した出力データを、前記ネットワークを介して他のサーバに送信し、
    　前記出力データを受信した他のサーバは、当該出力データに付加された前記識別フラグに基づき受信した出力データを前記マスタサーバに送信する、コンピュータシステム。     A computer system in which a master server and a slave server that executes an application according to input data transmitted from the master server are connected via a network,
    The master server periodically transmits a clear command for detecting a communication abnormality with the slave server to the slave server,
    The slave server initializes the watchdog timer in response to the clear command and starts a countdown, and when the watchdog timer times out, completes the processing of the application that has started execution before the timeout occurs, Sending output data with an identification flag indicating that the process has been completed after the watchdog timer has timed out to the process result of the completed application to another server via the network,
    The other server that has received the output data transmits the received output data to the master server based on the identification flag added to the output data.
  5. 　前記スレーブサーバは、前記ウォッチドックタイマのタイムアウト後、タイムアウト発生までに実行を開始しているアプリケーションの全ての処理が完了したことを示す通知信号を前記マスタサーバに送信することを特徴とする請求項４に記載のコンピュータシステム。 The slave server, after a timeout of the watchdog timer, transmits to the master server a notification signal indicating that all processes of an application that has started execution before the timeout occurs are completed. 5. The computer system according to 4.
PCT/JP2014/002435 2014-05-08 2014-05-08 Server device and computer system WO2015170365A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2014/002435 WO2015170365A1 (en) 2014-05-08 2014-05-08 Server device and computer system
TW103116947A TW201543224A (en) 2014-05-08 2014-05-14 Server device and computer system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/002435 WO2015170365A1 (en) 2014-05-08 2014-05-08 Server device and computer system

Publications (1)

Publication Number Publication Date
WO2015170365A1 true WO2015170365A1 (en) 2015-11-12

Family

ID=54392236

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2014/002435 WO2015170365A1 (en) 2014-05-08 2014-05-08 Server device and computer system

Country Status (2)

Country Link
TW (1) TW201543224A (en)
WO (1) WO2015170365A1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000270003A (en) * 1999-03-17 2000-09-29 Hitachi Telecom Technol Ltd Communication terminal device and control method for terminal monitor packet
JP2003122653A (en) * 2001-10-09 2003-04-25 Fujitsu Ltd Processing method related with time-out of server in client/server system and time-out processing selecting method and server computer and program therefor
US6615265B1 (en) * 1998-08-04 2003-09-02 International Business Machines Corporation Enabling planned outages of application servers
JP2004274275A (en) * 2003-03-06 2004-09-30 Canon Inc Communication system
JP2007199829A (en) * 2006-01-24 2007-08-09 Hitachi Ltd Computer control method, information processing system, operation management device, computer, and computer control program
JP2011191942A (en) * 2010-03-12 2011-09-29 Canon Inc Processing method and apparatus

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6615265B1 (en) * 1998-08-04 2003-09-02 International Business Machines Corporation Enabling planned outages of application servers
JP2000270003A (en) * 1999-03-17 2000-09-29 Hitachi Telecom Technol Ltd Communication terminal device and control method for terminal monitor packet
JP2003122653A (en) * 2001-10-09 2003-04-25 Fujitsu Ltd Processing method related with time-out of server in client/server system and time-out processing selecting method and server computer and program therefor
JP2004274275A (en) * 2003-03-06 2004-09-30 Canon Inc Communication system
JP2007199829A (en) * 2006-01-24 2007-08-09 Hitachi Ltd Computer control method, information processing system, operation management device, computer, and computer control program
JP2011191942A (en) * 2010-03-12 2011-09-29 Canon Inc Processing method and apparatus

Also Published As

Publication number Publication date
TW201543224A (en) 2015-11-16

Similar Documents

Publication Publication Date Title
US10999367B2 (en) Communication apparatus, control device, and communication method
JP2018038081A5 (en)
US20140078926A1 (en) Initiator apparatus, target apparatus, communication system, timeout detection method, and timeout detection program
US9442786B2 (en) Determining and correcting software server error conditions
JP6431197B2 (en) Snapshot processing methods and associated devices
CN108710597B (en) Method and system for optimizing MCU communication by using GPIO
CN101710288A (en) Upgrade processing method and system, upgrade server and upgrade device
JP2013246447A5 (en)
US20160345057A1 (en) Data processing system and data processing method
US8775500B2 (en) Distributed computing system having leader signaled agents to execute task after data acquisition completion
EP4120631A1 (en) Network connection method and device for training participant end of common training model
CN104410687A (en) Method for transmitting data among multiple groups of courses on basis of pipe
WO2015170365A1 (en) Server device and computer system
WO2016197693A1 (en) Operation instruction processing method and apparatus
JP6919461B2 (en) Node device and failover control method
JP2016218923A5 (en)
JP2016167203A (en) Information processing device and control method of information processing device
JP2005277978A (en) Method and device for automatically setting identification number
WO2016080112A1 (en) Control device
JP2015121971A5 (en) Information processing system, information processing apparatus, image forming apparatus, information processing method, and information processing program
CN110968022B (en) Programmable logic controller, programmable logic controller system, and control method
CN104714909B (en) Processing bus hangs dead device, method, bus structures and system
TWI793328B (en) Communication system, control system and communication device
US11954509B2 (en) Service continuation system and service continuation method between active and standby virtual servers
JP6962243B2 (en) Computer system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14891134

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14891134

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP