JP6611254B2 - Server apparatus, communication system, and communication method - Google Patents

Description

  The present invention relates to a server device, a communication system, and a communication method.

  Conventionally, a DB server system in which a DB is made redundant by a master database (DB) / slave DB is known (see Patent Document 1). The service providing apparatus usually provides a service to the user with reference to data in the master DB.

  In such a DB server system, the master DB and the slave DB cannot be synchronized due to a network failure or the like, and even if the so-called master / slave replacement to replace the master DB and the slave DB is not possible, the service can be provided. It is necessary to provide it continuously.

JP2013-142996A

  However, in a DB server system having a redundant configuration with a master / slave, if a network failure occurs between areas where the master and the slave are installed, there is a problem that the data update processing service is interrupted.

  Specifically, a conventional DB server system will be described with reference to FIG. FIG. 15 is a diagram for explaining a communication method of a conventional DB server system. As shown in FIG. 15, in the conventional DB server system, for example, a master DB server 20P is provided in the area A, and accommodates the service providing devices 10Ap and 10Bp in the area A. The service providing devices 10Ap and 10Bp are connected to users 5Ap to 5Bp and 5Cp to 5Dp, respectively. In the area B, the slave DB server 30P is provided and accommodates the service providing apparatuses 10Cp and 10Dp in the area B. The service providing apparatuses 10Cp and 10Dp are connected to users 5Ep to 5Fp and 5Gp to 5Hp, respectively.

  Among these, in the area A, even when a network failure occurs (see (1) in FIG. 15), the master DB server 20P can be accessed, so that the service providing apparatuses 10Ap and 10Bp perform processing for referring to data. In addition, processing for updating data can be executed.

  For example, when the service providing apparatus 10Ap receives a data update request from the user 5Ap (see (2-1) in FIG. 15) and transmits the data update request to the master DB server 20P (see (3-1) in FIG. 15). The master DB server 20P updates the data (see (4-1) in FIG. 15). When the service providing apparatus 10Ap receives a data reference request from the user 5Bp (see (2-2) in FIG. 15), the service providing apparatus 10Ap transmits a data reference request to the master DB server 20P ((3-2) in FIG. 15). Reference), a data reference response by the master DB server 20P can also be transmitted to the user 5Bp.

  On the other hand, in the area B, when a network failure occurs (see (1) in FIG. 15), the service providing apparatuses 10Cp and 10Dp can access the slave DB server 30P because only the slave DB server 30P can be accessed. Although the process of referring to the data can be performed, the process of updating the data cannot be performed.

  For example, even when the service providing apparatus 10Cp receives a data reference request from the user 5Fp (see (2-4) in FIG. 15), communication with the master DB server 20P is impossible due to a network failure ((( 4-3)). In this case, the service providing apparatus 10Cp accesses the slave DB server 30P (see arrow Yp in FIG. 15), transmits a data reference request (see (3-4) in FIG. 15), and slave DB server 30P. The data reference response by is transmitted to the user 5Fp.

  On the other hand, when receiving a data update request from the user 5Ep (see (2-3) in FIG. 15), the service providing apparatus 10Cp cannot communicate with the master DB server 20P due to a network failure (see FIG. 15). 15 (4-3)), the data update request cannot be transmitted to the master DB server 20P (see (3-3) in FIG. 15).

  As described above, in the conventional DB server system, the service providing apparatuses 10Cp and 10Dp in the area B that can only access the slave DB server 30P can access the slave DB server 30P and refer to the data, but update the data. Processing cannot be performed. As a result, there is a problem that in the area B where only the slave DB server 30P can be accessed, the data update processing service is interrupted (see (5) in FIG. 15).

  The present invention has been made in view of the above, and in a configuration having a redundant configuration by master / slave, a server device, a communication system, and a communication device for preventing interruption of data update processing service in an area due to a network failure An object is to provide a communication method.

  In order to solve the above-described problems and achieve the object, the server device according to the present invention is a server device connected to the opposite device via a network in a master or slave relationship, and data recorded in units of records If the device is a master and the device is a master, the data in the database is updated when a data update instruction is received from an external device. When the synchronization instruction is received from the device, the data update unit that updates the data in the database and the own device is the master, and the data update unit receives the data update instruction from the external device and When the data is updated, update data that transmits an instruction to synchronize with the updated data to the opposite device as a slave When a network failure occurs between the communication unit and the opposite device, a master transition unit that temporarily transitions its own device to the master and an external device that is provisionally transitioned to the master by the master transition unit And a provisional data update unit that updates data in the database asynchronously with the counterpart device when a data update instruction is received from the device.

  According to the present invention, in a configuration having a redundant configuration with master / slave, interruption of data update processing service in an area due to a network failure is prevented.

FIG. 1 is a schematic diagram showing a schematic configuration of a communication system according to the present embodiment. FIG. 2 is a schematic view illustrating the configuration of the service providing apparatus shown in FIG. FIG. 3 is a schematic view illustrating the configuration of the master DB server shown in FIG. FIG. 4 is a diagram showing an example of the data configuration of data stored in the master DB shown in FIG. FIG. 5 is a diagram for explaining the matching process of the matching unit shown in FIG. FIG. 6 is a schematic view illustrating the configuration of the slave DB server shown in FIG. FIG. 7 is a diagram illustrating an example of a normal process flow of the communication system illustrated in FIG. 1. FIG. 8 is a diagram illustrating an example of a processing flow when a network failure occurs in the communication system illustrated in FIG. 1. FIG. 9 is a diagram for explaining data update processing at the time of a network failure in the communication system shown in FIG. FIG. 10 is a diagram illustrating an example of the flow of data reference processing when a network failure occurs in the communication system illustrated in FIG. FIG. 11 is a diagram showing an example of the flow of matching processing for data after network recovery of the communication system shown in FIG. FIG. 12 is a sequence diagram showing a processing procedure of the communication system shown in FIG. FIG. 13 is a flowchart showing the procedure of the matching process shown in FIG. FIG. 14 is a diagram illustrating an example of a computer in which a master DB server or a slave DB server is realized by executing a program. FIG. 15 is a diagram for explaining a communication method of a conventional DB server system.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. Moreover, in description of drawing, the same code | symbol is attached | subjected and shown to the same part.

[Embodiment]
First, regarding the communication system according to the embodiment, a schematic configuration of the communication system, a flow of communication processing in the communication system, and a specific example will be described.

[Communications system]
FIG. 1 is a schematic diagram showing a schematic configuration of a communication system according to the present embodiment. As shown in FIG. 1, the communication system 1 includes service providing apparatuses 10A and 10B in area A, service providing apparatuses 10C and 10D in area B, a master database (DB) server 20 in area A, and a slave in area B. The DB server 30 is configured to be communicably connected to each other via a network. Users 5A to 5B, 5C to 5D, 5E to 5F, and 5G to 5H are connected to the service providing devices 10A to 10D, respectively.

  The service providing devices 10A to 10D provide services to connected users 5A to 5B, 5C to 5D, 5E to 5F, and 5G to 5H, respectively. The master DB server 20 has a master DB 21 that stores data necessary for service provision. The slave DB server 30 has a slave DB 31 that stores duplicate data of the master DB 21.

  In the communication system 1 having such a configuration, the service providing apparatuses 10A to 10D are normally connected to the masters 5A to 5B, 5C to 5D, 5E to 5F, and 5G to 5H, respectively, in advance or in advance. The data cache is accessed by accessing the DB 21 and referring to the data, and storing the referenced data in the cache memory. Then, the service providing apparatuses 10A to 10D provide services to the users 5A to 5B, 5C to 5D, 5E to 5F, and 5G to 5H, respectively, that are connected using the data. In addition, the service providing apparatuses 10A to 10D access the master DB 21 in response to requests from the users 5A to 5B, 5C to 5D, 5E to 5F, and 5G to 5H to connect to the master DB 21 in normal times. Update.

  In addition, when the service providing apparatuses 10C and 10D in the area B cannot communicate with the master DB server 20 due to a network failure or the like, the slave B 31 accesses the slave DB 31 to refer to data, performs data cache, and uses this data. A service is provided to connected users 5E to 5F and 5G to 5H. When a network failure occurs, the service providing apparatuses 10C and 10D access the slave DB 31 and update the data of the slave DB 31 in response to requests from the users 5E to 5F and 5G to 5H to be connected.

  The master DB server 20 receives a data update instruction and synchronizes data with the slave DB server 30. The master DB server 20 notifies the slave DB server 30 of the updated data and performs replication to synchronize the master DB 21 and the slave DB 31 when the data update of the master DB 21 is performed by an operation terminal (not shown) or the like. .

  When a network failure occurs between the master DB server 20 and the slave DB server 30, the master DB server 20 temporarily moves to the master and updates data in the master DB 21 asynchronously with the slave DB server 30. In other words, the master DB server 20 continues the data update process of the master DB 21 in a range closed to the area A of the service providing apparatuses 10A and 10B accommodated in the master DB server 20. That is, when a network failure occurs between the master DB server 20 and the slave DB server 30, the master DB server 20 does not execute replication with the slave DB server 30 that is the opposite device.

  Further, when the network failure is recovered, the master DB server 20 checks the data matching degree in units of records between the data held by the slave DB 31 and the data held by the master DB 21. Subsequently, when there is a difference between the data, the master DB server 20 determines correct data based on the history information of the data, and performs data matching with the slave DB server 30 by differential replication.

  The slave DB server 30 updates data in synchronization with the master DB server 20. The slave DB server 30 receives the updated data from the master DB server 20 and performs replication for synchronizing the slave DB 31 with the master DB 21. Then, when a network failure occurs between the slave DB server 30 and the master DB server 20, the slave DB server 30 temporarily transitions to the master and executes data update processing of the slave DB 31 asynchronously with the master DB server 20. In other words, the slave DB server 30 continues the data update process of the slave DB 31 in the range closed to the area B of the service providing apparatuses 10C and 10D accommodated in the slave DB server 30.

  In addition, when the network failure is recovered, the slave DB server 30 checks the data matching degree for each record of the data held by the slave DB 31 and the data held by the master DB 21. Subsequently, when there is a difference between the data, the slave DB server 30 determines correct data based on the history information of the data, and performs data matching with the master DB server 20 by differential replication.

  As described above, in the present embodiment, the master DB server 20 and the slave DB server 30 both temporarily move to the master when a network failure occurs. Both the master DB server 20 and the slave DB server 30 execute data update processing of each DB asynchronously with the opposing server. Therefore, in the present embodiment, the data update processing service can be continued even in the area B where only the slave DB server 30 can be accessed.

[Configuration of service providing device]
First, the configuration of the service providing apparatuses 10A to 10D will be described. FIG. 2 is a schematic view illustrating the configuration of the service providing apparatus 10C. The service providing apparatuses 10A, 10B, and 10D have the same functional configuration as the service providing apparatus 10C. Hereinafter, when the service providing apparatuses 10A to 10D are expressed without distinction, they are expressed as the service providing apparatus 10.

  The service providing device 10C is realized by a general-purpose computer such as a workstation or a personal computer, and an arithmetic processing device such as a CPU (Central Processing Unit) executes a processing program stored in a memory, as illustrated in FIG. Functions as the data reference unit 11 and the service providing unit 12.

  When the data reference unit 11 cannot communicate with the master DB server 20, the data reference unit 11 transmits a data reference or data update request of the slave DB 31 to the slave DB server 30 instead of the master DB server 20. Specifically, the data reference unit 11 transmits a request for data reference of the master DB 21 or a request for data update of the master DB 21 to the master DB server 20 in normal times. If the data reference unit 11 cannot access the master DB server 20 due to a network failure or the like, the data reference unit 11 transmits a data reference request or data update request to the slave DB 31 to the slave DB server 30.

  Further, the data reference unit 11 stores data included in the reference response returned from the master DB server 20 or the slave DB server 30 in response to the reference request in the cache memory (hereinafter referred to as a data cache).

  The service providing unit 12 provides services to the users 5E to 5F using the data cached data. For example, the service providing unit 12 provides a service for notifying the users 5E and 5F of the transfer number corresponding to the telephone number designated by the users 5E and 5F using the data in which the telephone number and the transfer number are associated with each other. .

[Master DB server configuration]
Next, the configuration of the master DB server 20 will be described. FIG. 3 is a schematic diagram illustrating the configuration of the master DB server 20. The master DB server 20 is realized by a general-purpose computer such as a workstation or a personal computer. As shown in FIG. 3, the master DB server 20 includes a master DB 21, an input unit 22, an output unit 23, a communication control unit 24, and a control unit 25. Prepare.

  The master DB 21 is realized by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The master DB 21 may be configured to communicate with a control unit 25 (described later) via a communication control unit 24 (described later).

  The master DB 21 stores service information used by the service providing apparatus 10 for providing services. FIG. 4 is a diagram illustrating an example of a data configuration of data stored in the master DB 21.

  As shown in the table T1 of FIG. 4, the service information includes a record ID for identifying each record, data indicating the telephone number and the information content of the transfer number that is the transfer destination of the telephone number, and data of these data. The time stamp indicating the update time is associated with the data update count indicating the number of times the data of this record has been updated.

  In the table T1 of FIG. 4, for example, the telephone number of the record ID “2” is “050-3333-4444”, and the transfer number is “090-3333-4444”. Further, for the record ID “2”, the time stamp is “03/17 11:35:10”, and the data update count is “14105”. The data portion of the service information is acquired from the upper OpS, the user, or the like under the control of the control unit 25 or in advance via the communication control unit 24 or the input unit 22.

  The input unit 22 is realized by using an input device such as a keyboard or a mouse, and inputs various instruction information such as processing start to the control unit 25 in response to an input operation by the operator. The output unit 23 is realized by a display device such as a liquid crystal display, a printing device such as a printer, an information communication device, or the like.

  The communication control unit 24 is realized by a NIC (Network Interface Card) or the like, and is connected to an external device such as a slave DB server 30 or an upper OpS (Operation System) via a telecommunication line such as a LAN (Local Area Network) or the Internet. Controls communication with the control unit 25.

  The control unit 25 includes an internal memory for storing a program that defines various processing procedures and the necessary data, and executes various processes using these programs. For example, the control unit 25 is an electronic circuit such as a CPU or MPU (Micro Processing Unit). As illustrated in FIG. 3, the control unit 25 includes a data update unit 251, an update data transmission unit 252, and a provisional master control unit 253.

  The data update unit 251 updates the data in the master DB 21 when update data is input from a higher level OpS or a user. Specifically, when data for updating service information is input via the communication control unit 24 or the input unit 22, the data update unit 251 updates the data of the corresponding record to the input data. When the data update unit 251 receives a data reference request from the service providing apparatuses 10A to 10D during normal times, the data updating unit 251 causes the service providing apparatuses 10A to 10D to refer to the data in the master DB 21. The data update unit 251 updates the data in the master DB 21 when receiving an update request from the service providing apparatuses 10A to 10D during normal times. The data update unit 251 responds to a data reference request by the service providing apparatuses 10A and 10B when a network failure occurs.

  The update data transmission unit 252 transmits the record with the updated data to the slave DB server 30 after the data in the master DB 21 is updated. Specifically, the update data transmission unit 252 includes the record ID corresponding to the updated data in the slave DB server 30 immediately or at a predetermined timing when the data update unit 251 updates the service information. Replication including the record is instructed to synchronize the master DB 21 and the slave DB 31.

  When a network failure occurs with the slave DB server 30, the temporary master control unit 253 tentatively shifts its own device to the master and updates data in the master DB 21 asynchronously with the slave DB server 30. The provisional master control unit 253 includes a master transfer unit 254, a provisional data update unit 255, a reliability notification unit 256, and a matching unit 257. When a network failure occurs with the slave DB server 30, the master migration unit 254 provisionally migrates to the master.

  When the provisional master control unit 253 tentatively transitions to the master, the provisional data update unit 255 receives the data update instruction from an external device and receives the data in the master DB 21 asynchronously with the opposite device. Update. That is, when a network failure occurs between the temporary data update unit 255 and the slave DB server 30, the temporary data update unit 255 responds to a data update request from the service providing apparatuses 10A and 10B accommodated in the master DB server 20 according to the data update request. Update the data in the DB 21.

  In this case, when the data is updated for each piece of data in the master DB 21 during the network failure continuation period, the provisional data update unit 255 associates the data with history information about the data update and stores the data in the master DB 21. As this history information, the provisional data update unit 255 stores a time stamp indicating the data update time and the number of data updates in the master DB 21. For example, in the above-described table T1 (see FIG. 4), when the transfer number of the record ID “2” is changed, the update time of the transfer number for the record ID “2” is recorded in the time stamp column, and the data Record the update count in the data update count column.

  The reliability notification unit 256, when requested to refer to data from the service providing apparatuses 10A and 10B, determines the reliability of the data requested to be referred to based on the history information, and is requested to refer to the data. Along with the data, the determined reliability is returned.

  Specifically, the reliability notification unit 256 determines that the reliability of the data requested for data reference is low when the number of updates of the data of the record ID requested for data reference is equal to or greater than a predetermined threshold. . This is because the data of this record ID may be updated even in the opposite slave DB server 30, and the data of the master DB 21 may not match the data of the slave DB 31. The threshold is set for each record in accordance with the control of the upper OpS and the like via the communication control unit 24 and the instruction information input from the input unit 22. The threshold value may be set in advance or may be periodically changed based on the statistics of the number of data updates so far.

  On the other hand, the reliability notification unit 256 determines that the reliability of the data requested for data reference is high when the number of updates of the data of the record ID requested for data reference is less than a predetermined threshold. This is because the data of the record ID is unlikely to be updated even in the opposing slave DB server 30, and the data of the master DB 21 and the data of the slave DB 31 are considered to be highly likely to match. .

  When the network failure is recovered, the matching unit 257 checks the data matching degree in units of records for the data held by the slave DB server 30 and the data held by the master DB 21. If there is a difference between the data whose degree of coincidence is confirmed, the matching unit 257 determines correct data based on the history information, and performs matching of the data with the slave DB server 30 by differential replication. For example, the matching unit 257 determines that, among the data held by the slave DB server 30 and the data held by the master DB 21, the data with the new time stamp for the determination target record is correct data.

  Specifically, the matching process of the matching unit 257 will be described with reference to FIG. FIG. 5 is a diagram for explaining the matching process of the matching unit 257 shown in FIG. In FIG. 5, at the time of network failure recovery, the data held by the master DB 21 is referred to as a table T1 ′, and the data held by the slave DB 31 is referred to as a table T2 ′. A case where the matching unit 257 performs matching processing for the record ID “2” will be described.

  As shown in FIG. 5, the matching unit 257 compares the data with the record ID “2” in the table T1 ′ and the data with the record ID “2” in the table T2 ′ (see (1) in FIG. 5). In this case, the data in the cell C11 of the table T1 ′ is the transfer number “090-9999-0000”. On the other hand, the data in the cell C21 of the table T2 ′ is the transfer number “090-7777-8888”. Therefore, the data of the record ID “2” is different between the table T1 ′ and the table T2 ′ (see (2) in FIG. 5).

  Therefore, the matching unit 257 determines which of the table T1 ′ and the table T2 ′ is correct data for the record ID “2” based on the history information. First, the matching unit 257 compares the time stamps of the table T1 ′ and the table T2 ′ for the record ID “2” (see (3) in FIG. 5).

  Then, the matching unit 257 determines that the data with the new time stamp is correct for the record ID “2”. In the example of FIG. 5, the time stamp of the table T2 ′ in the record ID “2” is “03/17 12:34:56” (see the cell C22). On the other hand, the time stamp of the table T1 ′ in the record ID “2” is “03/17 12:35:10” (see the cell C12), and it can be seen that the table T1 ′ is newer data. For this reason, in the example of FIG. 5, the matching unit 257 determines that the table T1 ′ having a new time stamp is correct data for the record ID “2” (see (4) of FIG. 5). As described above, the matching unit 257 performs data matching with the slave DB server 30 by differential replication after determining correct data based on the history information for each record.

[Configuration of slave DB server]
Next, the configuration of the slave DB server 30 will be described. FIG. 6 is a schematic view illustrating the configuration of the slave DB server 30. The slave DB server 30 is realized by a general-purpose computer such as a workstation or a personal computer. As shown in FIG. 6, the slave DB server 30 includes a slave DB 31, an input unit 32, an output unit 33, a communication control unit 34, and a control unit 35. Prepare. The slave DB server 30 and the master DB server 20 are, for example, similar server devices, and the master and slave functions can be switched between the slave DB server 30 and the master DB server 20.

  The slave DB 31 stores redundancy information obtained by duplicating the service information stored in the master DB 21. The input unit 32 is realized by using an input device such as a keyboard or a mouse, and inputs various instruction information such as processing start to the control unit 35 in response to an input operation by the operator. The output unit 33 is realized by a display device such as a liquid crystal display, a printing device such as a printer, an information communication device, or the like.

  The communication control unit 34 is realized by a NIC or the like, and controls communication between an external device such as the master DB server 20 and the service providing device 10 and the control unit 35 via an electric communication line such as a LAN or the Internet.

  The control unit 35 has an internal memory for storing a program that defines various processing procedures and necessary data, and executes various processes using these programs. For example, the control unit 35 is an electronic circuit such as a CPU or MPU. As illustrated in FIG. 6, the control unit 35 includes a reference data management unit 351, a data update unit 352, and a provisional master control unit 253.

  When receiving a data reference request from the service providing apparatuses 10C and 10D, the reference data management unit 351 causes the service providing apparatuses 10C and 10D to refer to the data in the master DB 21. In addition, when the master is absent, the reference data management unit 351 causes the service providing apparatuses 10C and 10D to refer to the redundancy information of the slave DB 31 when receiving a data reference request from the service providing apparatuses 10C and 10D.

  The data updating unit 352 updates the data in the slave DB 31 when receiving a synchronization instruction from the master DB server 20. That is, when the data update unit 352 receives a record with updated data from the master DB server 20, the data update unit 352 updates the data of the corresponding record in the slave DB 31 to the received data.

  In the same way as the temporary master control unit 253 in the master DB server 20, the temporary master control unit 253 tentatively transitions to the master when a network failure occurs with the opposing master DB server 20, and the master DB server 20. The data of the slave DB 31 is updated asynchronously. The provisional master control unit 253 includes a master transfer unit 254, a provisional data update unit 255, a reliability notification unit 256, and a matching unit 257, similar to the provisional master control unit 253 illustrated in FIG. Processing is executed according to 10D and the slave DB 31.

[Normal processing]
Here, the processing of the communication system 1 during normal times will be described. FIG. 7 is a diagram illustrating an example of a processing flow of the communication system 1 in a normal state.

  For example, when the service providing apparatus 10C in area B receives a data update request from the user 5E (see (1-1) in FIG. 7) and transmits the data update request to the master DB server 20 ((2-1) in FIG. 7). The master DB server 20 updates the data (see (3-1) in FIG. 7). Subsequently, the master DB server 20 notifies the updated data to the slave DB server 30, and performs replication to synchronize the master DB 21 and the slave DB 31 (see (4-1) in FIG. 7). The data update request processing by the area A service providing apparatuses 10A and 10B and the area B service providing apparatus 10D is executed in the same manner.

  When the service providing apparatus 10C receives a data reference request from the user 5F (see (1-2) in FIG. 7), the data providing request is transmitted to the master DB server 20 (see (2-2) in FIG. 7). Then, a data reference response by the master DB server 20 is transmitted to the user 5F. Alternatively, the service providing apparatus 10C transmits a data reference request to the slave DB server 30 when the master is absent (see (2-3) in FIG. 7), and transmits a data reference response by the slave DB server 30 to the user 5F. .

[Processing when a network failure occurs]
Next, processing of the communication system 1 at the time of network failure will be described. FIG. 8 is a diagram illustrating an example of a processing flow when the communication system 1 has a network failure.

  When a network failure occurs (see FIG. 8A) and communication between area A and area B becomes impossible, the master DB server 20 and the slave DB server 30 each move to a temporary master. That is, the master DB server 20 shifts from the master to the temporary master (see (1-1) in FIG. 8), and the slave DB server 30 shifts from the slave to the temporary master ((1-2 in FIG. 8). )reference). Then, the master DB server 20 and the slave DB server 30 continue the DB data update process in the closed areas A and B, respectively.

  Specifically, in the area A, for example, when the service providing apparatus 10A receives a data update request from the user 5B (see (2-1) in FIG. 8), the data update request is transmitted to the master DB server 20 ( (See (3-1) in FIG. 8). In response to this, the master DB server 20 updates the data (see (4-1) in FIG. 8). On the other hand, in area B, for example, when the service providing apparatus 10C receives a data update request from the user 5E (see (2-2) in FIG. 8), the data update request is transmitted to the slave DB server 30 (see FIG. 8). 8 (see 3-2)). In response to this, the slave DB server 30 updates the data (see (4-2) in FIG. 8).

  Therefore, with reference to FIG. 9, a data update process at the time of a network failure of the communication system 1 will be described. FIG. 9 is a diagram illustrating an example of the flow of data update processing of the communication system 1 when a network failure occurs. FIG. 9 also shows a table T1 ′ of the master DB 21 and a table T2 ′ of the slave DB 31. First, when a network failure occurs (see FIG. 9A) and communication between the area A and the area B becomes impossible, the master DB server 20 and the slave DB server 30 move to temporary masters, respectively. (Refer to (1-1) and (1-2) in FIG. 9).

  For example, in area B, the user 5E transmits a data update request for updating the transfer number of the record ID “2” from “090-3333-4444” to “090-7777-8888” to the service providing apparatus 10C (FIG. 9 (2)). In response to this, the service providing apparatus 10C transmits a data update request to the slave DB server 30 (see (3) in FIG. 9). In this case, since the slave DB server 30 functions as a provisional master, the slave DB 31 updates the data in the slave DB 31 in response to the data update request (see (4) in FIG. 9).

  As a result, the slave DB server 30 changes the record ID “2” transfer number to “090-7777-8888” (see cell C21 in FIG. 9). Then, the slave DB server 30 writes the data update time “03/17 12:34:56” in the time stamp column of the record ID “2” (see cell C22 in FIG. 9). Subsequently, the slave DB server 30 updates the data update count “14105” in the data update count column of the record ID “2” to “14106” (see cell C23 in FIG. 9). As a result, the data in the slave DB 31 is updated to the table T2 ′.

  In area A, the user 5B transmits a data update request for updating the transfer number of the record ID “2” from “090-3333-4444” to “090-9999-0000” to the service providing apparatus 10A (FIG. 9 (5)). In response to this, the service providing apparatus 10A transmits a data update request to the master DB server 20 (see (6) in FIG. 9). In this case, since the master DB server 20 functions as a provisional master, the data of the master DB 21 is updated in response to the data update request (see (7) in FIG. 9).

  As a result, the master DB server 20 changes the record ID “2” transfer number to “090-9999-0000” (see cell C11 in FIG. 9). Then, the master DB server 20 writes the data update time “03/17 12:35:10” in the time stamp column of the record ID “2” (see cell C12 in FIG. 9). Subsequently, the master DB server 20 updates the data update count “14105” in the data update count column of the record ID “2” to “14106” (see cell C13 in FIG. 9). As a result, the data in the master DB 21 is updated to the table T1 ′.

  Next, a data reference process at the time of network failure will be described with reference to FIG. FIG. 10 is a diagram illustrating an example of the flow of data reference processing when a network failure occurs in the communication system 1. In FIG. 10, an example of the data reference process in the area B will be described, but the data reference process is executed in the area A by performing the same process.

  As shown in FIG. 10, for example, in area B, the user 5F transmits a data reference request with the record ID “2” to the service providing apparatus 10C (see (8) in FIG. 10). In response to this, the service providing apparatus 10C transmits a data reference request with the record ID “2” to the slave DB server 30 (see (9) in FIG. 10).

  The slave DB server 30 receives this data reference request, and refers to the data of the record ID “2” from the table T2 ′ of the slave DB 31. Furthermore, the slave DB server 30 determines the reliability of the data with the record ID “2” based on the number of times the data is updated with respect to the record ID “2” and a threshold (for example, 10).

  In the table T2 ′, the data update count of the record ID “2” is “14106” (see the cell C23 of the table T2 ′), which is equal to or greater than the threshold value “10”, so it can be determined that the data reliability is low. Therefore, the slave DB server 30 responds to the service providing apparatus 10C with the telephone number “050-3333-4444” and the transfer number “090-7777-8888” of the record ID “2” and the record ID “2”. Is notified that the reliability of data reliability is “low” (see (10) in FIG. 10).

  Accordingly, the service providing apparatus 10C transmits a data reference response by the slave DB server 30 and a notification about the data reliability to the user 5F (see (11) in FIG. 10). As a result, the user 5F can recognize the reliability of the response data along with the content of the response data (see (12) in FIG. 10). In other words, the user can determine the reliability of the data referred to during the network failure period.

  Next, with reference to FIG. 11, the matching process for the data of the master DB 21 and the slave DB 31 after the network recovery will be described. FIG. 11 is a diagram illustrating an example of a flow of matching processing for data after the network recovery of the communication system 1.

  First, when the network is restored between the master DB server 20 and the slave DB server 30 (see (13) in FIG. 11), each checks a data difference (see (14) in FIG. 11). In this case, the master DB server 20 and the slave DB server 30 check the data matching degree in units of records for the data held by the master DB 21 and the data held by the slave DB 31.

  Specifically, the table T1 ′ of the master DB 21 and the table T2 ′ of the slave DB 31 are compared for each record. For example, for record ID “2”, the transfer number in table T1 ′ is “090-9999-0000” (see cell C11), whereas the transfer number in table T2 ′ is “090-7777-8888”. Different (see cell C21). Further, the table T1 ′ and the table T2 ′ have the same data update count. Therefore, the time stamps of the table T1 ′ and the table T2 ′ are compared.

  In the example of FIG. 11, the time stamp of the table T2 ′ is “03/17 12:34:56” (see cell C22), whereas the time stamp of the table T1 ′ is “03/17 12:35:10”. (See cell C12) and the table T1 'is newer. Therefore, the master DB server 20 and the slave DB server 30 determine that the table T1 ′ is correct data for the record ID “2” (see (14) in FIG. 11). Then, the master DB server 20 and the slave DB server 30 perform data replication with the slave DB server 30 by performing differential replication on both sides (see (15) in FIG. 11). Thereafter, the master DB server 20 returns from the temporary master to the master, and the slave DB server 30 returns from the temporary master to the slave.

[Processing procedure of communication system]
Next, a processing procedure of the communication system 1 will be described with reference to FIG. FIG. 12 is a sequence diagram illustrating a processing procedure of the communication system 1.

  As shown in FIG. 12, when the master DB server 20 and the slave DB server 30 detect that a failure has occurred in the network (steps S1-1 and S1-2), the master DB server 20 and the slave DB server 30 respectively transition to the temporary master (step S2−). 1 and step S2-2). Thereafter, the master DB server 20 and the slave DB server 30 continue the DB data update process in the closed areas A and B, respectively.

  For example, the service providing apparatus 10A in area A receives a data update request from the user 5B (step S3-1), and transmits the data update request to the master DB server 20 (step S4-1). The master DB server 20 determines whether or not there is a data update request (step S5-1). When it is determined that there is a data update request (step S5-1: Yes), the master DB server 20 updates the data in the master DB 21 (step S6-1).

  Further, for example, the service providing apparatus 10A receives a data reference request from the user 5A (step S7-1), and transmits the data reference request to the master DB server 20 (step S8-1). The master DB server 20 determines whether there is a data reference request (step S9-1). If the master DB server 20 determines that there is a data reference request (step S9-1: Yes), the update count of the data requested to be referenced is less than the threshold value or the data requested to be referenced. It is determined whether the number of updates is equal to or greater than a predetermined threshold (step S10-1).

  When the master DB server 20 determines that the number of updates of data requested to be referenced is less than a predetermined threshold (step S10-1: less than the threshold), the master DB server 20 determines that the reliability of this data is high (step S11-). 1). On the other hand, when the master DB server 20 determines that the update count of the data requested to be referenced is equal to or greater than a predetermined threshold (step S10-1: equal to or greater than the threshold), the master DB server 20 determines that the reliability of the data is low (step S10-1). S12-1).

  Then, the master DB server 20 responds to the service providing apparatus 10A with the requested data and notifies the reliability of the data (step S13-1). In response to this, the service providing apparatus 10A transmits a data reference response by the master DB server 20 and a notification about the data reliability to the user 5A (step S14-1).

  Subsequently, the master DB server 20 determines that there is no data update request (step S5-1: No), determines that there is no data reference request (step S9-1: No), or data reference response. After the processing, it is determined whether or not the network has been restored (step S15-1). When the master DB server 20 determines that the network has not been restored (step S15-1: No), the master DB server 20 returns to step S5-1.

  On the other hand, in the area B, the slave DB server 30 performs the same processing as the master DB server 20 in order to shift to the temporary master. That is, when the service providing apparatus 10C in area B receives a data update request from the user 5E (step S3-2) and transmits a data update request to the slave DB server 30 (step S4-2), the slave DB server 30 It is determined that there is a data update request (step S5-2: Yes), and the data in the slave DB 31 is updated (step S6-2).

  For example, when the service providing apparatus 10C receives a data reference request from the user 5F (step S7-2) and transmits a data reference request to the slave DB server 30 (step S8-2), the slave DB server 30 It is determined that there is a reference request (step S9-2: Yes), and a determination is made regarding the number of times the data requested to be referenced has been updated (step S10-2).

  Then, when the slave DB server 30 determines that the number of updates of data requested to be referenced is less than a predetermined threshold (step S10-2: less than the threshold), the slave DB server 30 determines that the reliability of the data is high (step S10-2). S11-2). On the other hand, if the slave DB server 30 determines that the update count of the data requested to be referenced is equal to or greater than a predetermined threshold (step S10-2: equal to or greater than the threshold), the slave DB server 30 determines that the reliability of the data is low (step S10-2). S12-2).

  Then, the slave DB server 30 responds to the service providing apparatus 10C with the requested data and also notifies the reliability of the data (step S13-2), and the service providing apparatus 10C receives the data reference response and the data reliability. A notification about the degree is transmitted to the user 5F (step S14-2).

  Subsequently, the slave DB server 30 determines that there is no data update request (step S5-2: No), determines that there is no data reference request (step S9-2: No), or data reference response. After the processing, it is determined whether or not the network has been restored (step S15-2). If the slave DB server 30 determines that the network has not been restored (step S15-2: No), the slave DB server 30 returns to step S5-2.

  When the master DB server 20 and the slave DB server 30 determine that the network has been restored (step S15-1: Yes, step S15-2: Yes), the opposite device is detected (step S16) and synchronization is performed. The matching process is performed while taking it (steps S17-1 and S17-2). The master DB server 20 and the slave DB server 30 perform data matching by differential replication (step S18).

  In this way, the master DB server 20 and the slave DB server 30 perform differential replication in which only the data with the difference is replicated in both, and only the data with the difference is matched. The alignment processing time can be shortened. Subsequently, the master DB server 20 returns from the temporary master to the master (step S19-1), and the slave DB server 30 returns from the temporary master to the slave (step S19-2).

[Alignment processing]
Next, the processing procedure of the matching process (step S17-1) shown in FIG. 12 will be described. FIG. 13 is a flowchart showing the procedure of the matching process shown in FIG.

  As shown in FIG. 13, in the master DB server 20, the matching unit 257 first initializes the record ID number “n” to be matched and sets n = 1 (step S21). Subsequently, the matching unit 257 compares the data with the counter device (step S22). Specifically, the matching unit 257 compares the data of the record ID “n” of the master DB 21 with the data of the record ID “n” of the slave DB 31 of the slave DB server 30 facing the master DB server 20.

  Then, the matching unit 257 determines whether the data of the record ID “n” of the master DB 21 is the same as the data of the record ID “n” of the slave DB 31 (Step S23). When the matching unit 257 determines that the data of the compared record ID “n” is the same (step S23: Yes), the matching unit 257 maintains the data of the record ID “n” (step S24).

  On the other hand, when the matching unit 257 determines that the data of the record ID “n” of the master DB 21 and the data of the record ID “n” of the slave DB 31 are not the same (step S23: No), the record ID “n” of the master DB 21 And the time stamp of the record ID “n” of the slave DB 31 are compared (step S25). The matching unit 257 determines that the data with the newer time stamp is correct data (step S26), and records the data determined to be correct in the record “n” (step S27).

  Then, the matching unit 257 determines whether n = N (step S28). Here, “N” is the maximum value of the record ID number “n”. If the matching unit 257 determines that n = N is not satisfied (step S28: No), the matching unit 257 sets n = n + 1 (step S29) to perform data matching for the next record ID, and returns to step S22. On the other hand, when the matching unit 257 determines that n = N (step S28: Yes), the matching process ends. The slave DB server 30 can also execute the matching process (step S17-2 in FIG. 12) by performing the same processing procedure.

[Effect of the embodiment]
Thus, in this embodiment, both the master DB server 20 and the slave DB server 30 tentatively move to the master when a network failure occurs. Then, when the master DB server 20 and the slave DB server 30 are temporarily transferred to the master, when receiving a data update instruction from an external device, the master DB server 20 and the slave DB server 30 are both asynchronous with the opposite server. Update the data.

  Therefore, in the present embodiment, the data update processing service can be continued even in the area B where only the slave DB server 30 can be accessed. In other words, according to the present embodiment, in a configuration having a redundant configuration by master / slave, it is possible to prevent interruption of the data update processing service in the area due to a network failure, so even when the network failure is prolonged. The service can be provided.

  In this embodiment, when the master DB server 20 and the slave DB server 30 that have been transferred to the temporary master due to a network failure receive a data reference request, the reliability of the data is determined based on the number of times the data is updated. In addition to the data, the user also responds to the reliability of the data. Therefore, according to the present embodiment, the reliability of the data referred to during the network failure period can be determined on the user side.

  In this embodiment, data check is performed in units of records after network recovery. In the present embodiment, when there is a difference, the master DB server 20 or the slave DB server 30 determines which one holds the correct data based on the time stamp that is history information, and correct data. Has been rewritten. For this reason, even if there is a difference in the data held by the master DB server 20 and the slave DB server 30 after the network recovery, it can be matched with correct data.

  In the present embodiment, after the network is restored, the master DB server 20 and the slave DB server 30 perform differential replication in which only data having a difference is replicated in both. In other words, since the master DB server 20 and the slave DB server 30 match only data having a difference, the matching processing time can be shortened as compared with the case of matching all data. That is, according to the present embodiment, it is possible to provide a service even when a network failure is prolonged, and it is possible to perform service recovery appropriately and in a short time.

[Other embodiments]
[System configuration, etc.]
Each component of each illustrated device is functionally conceptual and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or a part of the distribution / integration is functionally or physically distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. Further, all or a part of each processing function performed in each device can be realized by a CPU and a program that is analyzed and executed by the CPU, or can be realized as hardware by wired logic.

  In addition, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually All or a part of the above can be automatically performed by a known method. In addition, the processing procedure, control procedure, specific name, and information including various data and parameters shown in the above-described document and drawings can be arbitrarily changed unless otherwise specified.

[program]
FIG. 14 is a diagram illustrating an example of a computer in which the master DB server 20 or the slave DB server 30 is realized by executing a program. The computer 1000 includes a memory 1010 and a CPU 1020, for example. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to the display 1130, for example.

  The hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, a program that defines each process of the master DB server 20 or the slave DB server 30 is implemented as a program module 1093 in which a code executable by the computer 1000 is described. The program module 1093 is stored in the hard disk drive 1090, for example. For example, a program module 1093 for executing processing similar to the functional configuration in the master DB server 20 or the slave DB server 30 is stored in the hard disk drive 1090. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

  The setting data used in the processing of the above-described embodiment is stored as program data 1094 in, for example, the memory 1010 or the hard disk drive 1090. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1090 to the RAM 1012 as necessary, and executes them.

  The program module 1093 and the program data 1094 are not limited to being stored in the hard disk drive 1090, but may be stored in, for example, a removable storage medium and read out by the CPU 1020 via the disk drive 1100 or the like. Alternatively, the program module 1093 and the program data 1094 may be stored in another computer connected via a network (LAN, WAN, etc.). The program module 1093 and the program data 1094 may be read by the CPU 1020 from another computer via the network interface 1070.

  Although the embodiment to which the invention made by the present inventor is applied has been described above, the present invention is not limited by the description and the drawings that form part of the disclosure of the present invention according to this embodiment. That is, other embodiments, examples, operation techniques, and the like made by those skilled in the art based on the present embodiment are all included in the scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Communication system 5A-5H User 10, 10A-10D Service provision apparatus 11 Data reference part 12 Service provision part 20 Master database (DB) server 21 Master DB
22, 32 Input unit 23, 33 Output unit 24, 34 Communication control unit 25, 35 Control unit 30 Slave DB server 31 Slave DB
251 and 352 Data update unit 252 Update data transmission unit 253 Temporary master control unit 254 Master transfer unit 255 Temporary data update unit 256 Reliability notification unit 257 Matching unit 351 Reference data management unit

Claims (5)

A server device connected to the opposite device via a network in a master or slave relationship,
A database having data recorded in record units;
When the own device is the master, the data in the database is updated upon receiving a data update instruction from an external device. When the own device is the slave, the opposite device that is the master A data update unit that updates the data in the database when receiving a synchronization instruction from
When the own device is the master and the data update unit accepts a data update instruction from an external device and the data in the database is updated, an instruction to synchronize with the updated data is issued. An update data transmission unit for transmitting to the opposite device as the slave;
When a network failure occurs with the opposite device, a master transition unit that provisionally migrates its own device to the master,
When the master migration unit tentatively migrates to the master, when receiving a data update instruction from an external device, the data in the database is updated asynchronously with the opposite device, and a network failure continuation period and when said updated data for each data in the database, provisional data updating unit that associates the history information about the data update to the data Ru is stored in the database,
When a data reference is requested from an external device, the reliability of the data requested to be referred to is determined based on the history information, and the reliability to which the reliability determined with the data requested for data reference is returned A degree notification section;
Have
The provisional data update unit stores the update number of data in the database as the history information,
The reliability notification unit determines that the reliability of the data requested for data reference is low when the number of updates is equal to or greater than a predetermined threshold, and the data when the number of updates is less than a predetermined threshold. A server device that judges that the reliability of data requested to be referenced is high . If network failure is recovered, the data of the data and the database that the opposing device held have to check the coincidence of the data record-by-record basis, based on the history information if there is a difference between the data 2. The server according to claim 1, further comprising: a matching unit configured to match the data with the opposite apparatus by differential replication in which correct data is determined and only data having a difference is replicated in both. apparatus. The provisional data update unit stores a time stamp indicating a data update time in the database as the history information,
The server device according to claim 2, wherein the matching unit determines that data having a new time stamp is correct among data held by the opposite device and data held by the database. A communication system having a master database server and a slave database server connected to the master database server via a network,
The master database server is
A master database having data recorded in record units;
A first data updating unit that updates data in the master database when receiving an instruction to update data from an external device;
When the data update instruction is received from the external device by the first data update unit and the data in the master database is updated, the synchronization instruction with the updated data is transmitted to the slave database server. An update data transmission unit to perform,
A first master migration unit that provisionally migrates its own device to a master when a network failure occurs with the slave database server;
Wherein when the first of the tentatively shifts to the master by the master transition, when accepting the instruction data update from an external device, and updates the data of the master database asynchronously with the slave database server , if the network failure duration was updated data for each data of the master database, the first interim data updating unit that associates the history information about the data update to the data Ru is stored in the master database,
When an external device requests data reference, the reliability of the data requested for data reference is determined based on the history information, and the reliability determined with the data requested for data reference is returned. 1 reliability notification unit;
Have
The first provisional data update unit stores the number of data updates as the history information in the master database,
The first reliability notification unit determines that the reliability of the data requested for data reference is low when the number of updates is equal to or greater than a predetermined threshold, and the number of updates is less than the predetermined threshold It is determined that the reliability of the data requested for the data reference is high,
The slave database server is
A slave database having data recorded in record units;
A second data updating unit that updates data in the slave database when receiving a synchronization instruction from the master database server;
A second master migration unit that provisionally migrates its own device to a master when a network failure occurs with the master database server;
Wherein when it is migrated to a second tentative master by the master transition, when accepting the instruction data update from an external device, and updates the data of the slave database asynchronously with the master database server , if the network failure duration was updated data for each data of the slave database, the second provisional data updating unit that associates the history information about the data update to the data Ru is stored in the slave database,
When an external device requests data reference, the reliability of the data requested for data reference is determined based on the history information, and the reliability determined with the data requested for data reference is returned. 2 reliability notification units;
Have a,
The second provisional data update unit stores the number of data updates in the slave database as the history information,
The second reliability notification unit determines that the reliability of the data requested for data reference is low when the update count is equal to or greater than a predetermined threshold, and the update count is less than the predetermined threshold And determining that the data requested to be referred to is highly reliable . A communication method executed by a server device connected to a counter device via a network in a master or slave relationship,
The server device has a database having data recorded in units of records,
When the own device is the master, the data in the database is updated upon receiving a data update instruction from an external device. When the own device is the slave, the opposite device that is the master A data update step of updating the data in the database when receiving a synchronization instruction from
If the device itself is the master, and the data update instruction is received from an external device in the data update process and the data in the database is updated, an instruction to synchronize with the updated data is issued. An update data transmission step for transmitting to the opposite device as the slave;
When a network failure occurs with the opposite device, a transition step of tentatively transferring the own device to the master,
When transitioning to the master provisionally by the migration step, when receiving a data update instruction from an external device, the data in the database is updated asynchronously with the opposite device, and the network failure continuation period is reached. when updating the data for each data of said database, and provisional data updating process in association with history information about the data update to the data Ru is stored in the database,
When a data reference is requested from an external device, the reliability of the data requested to be referred to is determined based on the history information, and the reliability to which the reliability determined with the data requested for data reference is returned Degree notification process,
Including
In the provisional data update step, as the history information, the number of data updates is stored in the database,
The reliability notification step determines that the reliability of the data requested for data reference is low when the number of updates is equal to or greater than a predetermined threshold, and the data when the number of updates is less than a predetermined threshold. A communication method characterized by determining that the reliability of data requested to be referenced is high .

Images