JP5366177B2 - Slot interface access device, method and program thereof, and redundant configuration and alternative method of main device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow a single main device to manage information such as hardware resources of all main devices connected to a network. <P>SOLUTION: A slot management module, a slot control module and a physical slot/management slot contrast table are provided between an input/output control module and its lower slot interface. The input/output control module accesses the slot interface using virtual slot identification information. Referring to the physical slot/management slot contrast table, the slot management module converts the virtual slot identification information to physical slot identification information, and accesses the slot control module corresponding to the physical slot identification information acquired by conversion, thereby achieving physical access to the slot interface by the input/output control module. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to a slot interface access apparatus, a method thereof, and a program thereof for accessing slot interfaces existing distributed in a plurality of main apparatuses.

  Further, the present invention provides a redundant configuration of a main device in which a plurality of main devices are installed and the master device serving as a master malfunctions, and another master device serves as an alternative master main device, and a method for replacing the main device About.

Conventionally, there has been a technology in which main devices are connected by a network and the functions of each other are used between the main devices. Here, the main device is a device having an interface for accommodating a terminal (such as a button telephone), an interface for connecting to a public line, or an interface for connecting to an IP network. is there.
International Publication No. 1997/35255 Pamphlet JP 2001-35836 A

  However, in order to make the functions of the main device available to other main devices via the network, it is necessary to modify each function. For this reason, functions that have not been modified are not compatible with the network, and it has been impossible to use all the functions of a certain main device via another network.

  In other words, in the past, networking connections between main units are such that each main unit manages resources with its own CPU and also manages the status of terminals, lines, etc. separately, so that the functions of the main unit are moved between networks. However, it was not easy to operate the function with a single main device, and it was necessary to redesign the function to cope with the network.

  In the conventional networking system, the slot of the package, which is the resource of the main device, is managed separately for each system, so that each system can know the information, status, etc. of each other's resources. Therefore, there has been a limitation when using the function of the main device on the network.

  Therefore, the present invention makes it easy to manage information and constructs a network without function restriction by centrally managing information such as hardware resources of all the main devices connected to the network by a single main device. For the purpose.

  In the present invention, all the resource information that was conventionally managed by each main device is centrally managed by one main device, and call control is also performed collectively by one main device. Avoid problems.

  In other words, the monitoring and control of the terminals, lines, etc. of other main devices connected to the network are all performed by a single main device so that the resources on the network are connected to the main device. Can be treated as if

  In other words, since the terminals and lines of other systems can be handled like their own terminals and lines, there is no need to remodel the main unit function to be compatible with the network as in the past, and it is even necessary to be aware of the network. Disappear.

  All functions including those not currently supported by the network can be used on the network.

  Further, according to the present invention, the above-mentioned problem is avoided by a single main device managing the slots of all the main devices connected to the network.

  That is, the master main unit that controls all call processing on the network centrally manages its own slot and the slot of the slave main unit connected to itself, even if it is a slot of another system, Treat as if

  By doing this, the master can manage all the information on the slot on the network, and at the same time, all resources such as terminals and lines connected to the package installed in the slot are also of the local system. Can be handled.

According to the present invention, a slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and a lower slot interface, and the input / output control module includes virtual slot identification information. The slot management module refers to the physical slot / management slot comparison table, converts the virtual slot identification information into physical slot identification information, and obtains the physical slot identification obtained by the conversion. By accessing the slot control module corresponding to the information, physical access to the slot interface by the input / output control module is realized , and if a new slot interface is added, a new virtual slot The And pressurizing, features and identification information of the slot control module corresponding to the added slot interface, further comprising means for adding the correspondence between the identification information of the added virtual slot the physical slot / managed slot comparison table A slot interface access device is provided.

According to the present invention, a slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and a slot interface below the input / output control module. The slot interface is accessed using the identification information, the slot management module refers to the physical slot / management slot comparison table, converts the virtual slot identification information into physical slot identification information, and obtains the physical obtained by the conversion. By accessing the slot control module corresponding to the slot identification information, physical access to the slot interface by the input / output control module is realized , and if a new slot interface is added, a new Virtual slot Add the door, and identification information of the slot control module corresponding to the added slot interface, further comprise a procedure for adding a correspondence between identification information added virtual slots on the physical slot / managed slot comparison table A slot interface access method is provided.

Further, according to the present invention, when a plurality of main devices are provided, one of the main devices is a master main device, the other main device is a slave main device, and the master main device becomes malfunctioning, Ri Do one slave main unit a new master main device among the slave main device, each main unit is assigned a priority, when the current master main device becomes dysfunctional, other The main device with the highest priority among the main devices is assumed to be the new master main device, and each main device transmits a priority inquiry signal to all other main devices in a brute force manner. The priority obtained by reply is compared with the priority of the main device itself. If the priority of the main device itself is higher than the priority of all other main devices, the main device itself becomes the master main device. In response to an inquiry from the main device If the device itself notifies that it is the master main device, and each main device receives a notification from the new main device that the main device is the master main device, the main device itself is a slave main device redundancy of the main device is provided, wherein the recognized as.

  Further, according to the present invention, in the redundant configuration of the main device, the new master main device monitors whether or not the master master device that has failed has recovered, and the master master that has failed. When the device recovers, a new master main device is provided with a redundant configuration of the main device characterized in that the entire master system is returned to the recovered master main device.

  Further, according to the present invention, when a plurality of main devices are provided, one of the main devices is a master main device, the other main device is a slave main device, and the master main device becomes malfunctioning, An alternative method of the main device is provided, wherein one of the slave main devices becomes a new master main device.

  Further, according to the present invention, in the above main device alternative method, the new master main device monitors whether or not the master master device that has failed has recovered, and the master master that has failed. When the device recovers, an alternative method of the main device is provided in which the new master main device returns control of the entire system to the recovered master main device.

Furthermore, according to the present invention, a slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and a lower slot interface, and the input / output control module includes a virtual slot. The slot interface is accessed using the identification information, the slot management module refers to the physical slot / management slot comparison table, converts the virtual slot identification information into physical slot identification information, and obtains the physical obtained by the conversion. by accessing slot control module corresponding to the slot identification information, a slot interface access apparatus characterized by physical access to the slot interface by said output control module are implemented, a new slot If an interface is added, a new virtual slot is added, and the correspondence between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is the physical slot / management. A slot interface access device is provided , further comprising means for adding to the slot comparison table, wherein the device has a higher CPU capacity than other devices having a slot interface.

Furthermore, according to the present invention, a slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and a lower slot interface, and the input / output control module includes a virtual slot. The slot interface is accessed using the identification information, the slot management module refers to the physical slot / management slot comparison table, converts the virtual slot identification information into physical slot identification information, and obtains the physical obtained by the conversion. by accessing slot control module corresponding to the slot identification information, a slot interface access method characterized by physical access to the slot interface by said output control module are implemented, a new slot If an interface is added, a new virtual slot is added, and the correspondence between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is the physical slot / management. There is further provided a procedure for adding to the slot comparison table, and the method is provided by a device having a higher CPU capacity than other devices having a slot interface.

  According to the present invention, all the theoretical functions of the main device can be used on the network. In the existing technology, each main device basically operates independently, and in order to move them in a coordinated manner, a special change is required for each function. However, according to the present invention, by introducing a mechanism for absorbing differences between networks in a lower layer, the functions of the upper layer can move without being aware of between networks.

  In addition, by using the present invention, a highly reliable networking can be constructed. In the conventional method, each main apparatus operates independently, and resources are managed separately. For this reason, there may be a problem in that there is a discrepancy in the state between resources depending on the timing. The present invention can also solve this problem.

  Furthermore, in the present invention, since the function is constructed in the lowest layer of the functions of the main apparatus, if even the lower layer is created without any problem, there is no functional failure in the upper layer, and centralized management of resources is performed. Thus, there is no inconsistency in the state between resources.

  Furthermore, by using the present invention, it is possible to construct a mechanism that makes networking unconscious.

  Furthermore, by introducing a mechanism that absorbs network differences at the lowest processing layer, higher-level applications can use functions without being aware of the network.

  This eliminates the need for network-compatible processing that was conventionally required at the application level, greatly reducing the number of network-compatible man-hours and improving the quality.

  As a result, the enormous man-hours required to modify the function for the conventional network can be reduced to zero, and the quality can be improved.

  The best mode for carrying out the present invention will be described below in detail with reference to the drawings.

[Embodiment 1]
The point of this embodiment lies in the technology of handling resources on the network like own resources.

  In the main apparatus that operates under program control, hardware resource management, that is, management of terminals, lines, etc., is performed in the form of package management.

  Therefore, in order to treat a resource on the network as if it were its own resource, it is only necessary to treat a package on the network as if it were its own package.

  FIG. 1 shows a conceptual diagram of package management between networks.

  When a package is installed in the main apparatus 2, the package information and information about terminals and lines connected thereto are transmitted to the main apparatus 1 via Ethernet (registered trademark).

  On the other hand, since the main device 2 does not transmit any of the information to the package control unit or the call control unit, the main device 2 does not seem to have changed the situation.

  The main device 1 processes the data coming from the main device 2 in a lower layer and makes it appear as if information has come from its own slot. Therefore, the main device 1 has a package stuck in its own slot. Looks like.

  In addition, an instruction for the package (downlink data) is also processed in a lower layer, and the instruction for the virtual package is transmitted to the actual package on the network.

  By introducing the above mechanism, it is possible to treat resources on the network as if they were own resources.

  Therefore, in higher layers such as call control, resources can be freely used without being conscious of being on the network.

  FIG. 2 shows a block diagram of networking according to the present invention.

  A master device that manages all resources on the network and performs all call control is called a master.

  On the other hand, the main device that connects to the master, provides package information, and follows the instructions from the master is called a slave.

  In order to perform networking according to the present embodiment, it is necessary for one main apparatus to become a master among a plurality of main apparatuses constituting the network. All the slaves connect to the master and follow the instructions, and do not perform any processing such as call control. That is, even if the slave has a function unit for performing call control or the like, these are in a dormant state.

  The master can control a plurality of slaves, and can handle all the resources of the main device connected as slaves as if it were its own.

  As described above, the network constituted by the master and the slave can behave as if it is one system.

  Information on which main device is to be a master or slave, and information on which IP address is connected to each must be set in advance.

  The main device set as the master waits for connection of the slave, and the slave connects to the IP address of the master set in advance.

  In this way, after the connection between the master and the slave is established, the package information is transmitted, and the network operates as a network.

  Also, if the master goes down, all main devices connected to it become unusable. To prevent this, when the master goes down, one of the one or more slaves becomes the master, acting as the master (Redundancy).

  Information about which slave will act when the master goes down needs to be set in advance.

  Next, a specific method for centrally managing resources on the network will be described.

  FIG. 3 shows a system configuration in networking.

  There is only one master on the network and it controls all slaves.

  In order to identify each system on the network, each system has a unique value called a system ID.

  FIG. 4 is a conceptual diagram of slot management in the present invention.

  Each system connected to the network and having a system ID has a package physically installed in the slot, but these pieces of information are centrally collected as a virtual slot database and managed by the master system. .

  The master refers to this virtual slot database to control the slot.

  If it is a slot of another system, the physical slot exists in the remote place connected by the IP network, but the master is not aware of the physical slot in the remote place, and is as if it is its own slot. Can be treated as if

  Therefore, the terminal and line connected to the package installed in the slot can be handled in the same way as the terminal and line connected to the own system.

  A diagram showing this is shown in FIG.

  A system ID: 1 system has a terminal connection package; a system ID: 2 system has a package accommodating a line connected to a public line; and a system: 3 system has an IP line connected to an IP network. Each package is installed.

  Since these physical slots are managed as virtual slots, the terminals, lines, etc. accommodated in the packages connected to the slots can be freely controlled as in the own system.

  According to the above method, the function of the main apparatus can be used without limitation even in a system distributed in a network.

  As shown in FIG. 3, each system shown in FIG. 5 has a server / client type configuration in which one master controls the other slaves. Performs device call processing and database management. The master also manages the virtual slots.

  Each system is connected by IP and has a unique system ID for identifying each system.

  The system 1 accommodates a package for accommodating a terminal, the system 2 accommodates a package for accommodating a normal line, and the system 3 accommodates a package for accommodating an IP line.

  Information is managed in the virtual slot database. This data is basically managed by the master, but the same data is held in each slave in case the master changes.

  The embodiment shown in FIG. 5 will be described from the viewpoint of data flow.

  FIG. 6 shows the data flow of conventional package control.

  As shown in FIG. 6, the upstream data from the package is sent from the slot I / F module 101 to the CAPS (call control module) / OPMS (package, terminal management module) 105 via the IOCS (input / output control module) 103. It is done.

  Then, the data is processed by the CAPS / OPMS 105, and a downlink instruction is issued to the slot I / F module 101 via the IOCS 103. For example, when a package is installed, data is sent to the OPMS 105 as upstream data, and it is recognized that the package has been installed. Even when the terminal connected to the package is off-hooked, the CAPS 105 is notified as upstream data to the CAPS 105, and the CAPS 105 receives it and sends a command to the terminal to output a dial tone as downstream data via the IOCS 103. , Etc.

  In FIG. 6, since the data from the slot I / F 101 is directly input to the upper module as input data, naturally, only the slot connected to the own system can be controlled.

  FIG. 7 shows a data flow in this embodiment.

  As shown in FIG. 7, slot management module 107 for controlling slot input / output and slot management module 109 for managing slot information are newly added to realize slot management via the network.

  Upstream data from the slot is once spooled in the slot control module 107 and sent to the slot management module 109 of the master to which the own system belongs. If it is the master, it is sent to its own slot management module 109. The data sent to the slot management module 109 makes the data appear to the IOCS 103, which is a higher module, as if the data has come up from a specific slot.

  Here, the operation of the slot management module 109 will be described with reference to FIG.

  When the slot management module 109 receives data from a specific slot of a specific system, if it is a slot of a system that has not been recognized so far, a new virtual slot number is assigned. Thereafter, the slot of the system is treated as the slot of the assigned virtual slot number.

  For example, when uplink data comes up from slot 1 of system 1, if it is a slot that has not been recognized so far, slot 1 is assigned as a virtual slot.

  In this way, when a new virtual slot is assigned, a physical slot / virtual slot comparison table 111 as shown in FIG. 8 is created.

  Thereafter, the data coming up from the slot 1 of the system 1 is handled as if it is the slot 1 of its own system in the higher modules such as IOCS, CAPS, etc., so that it is not necessary to be aware of the network.

  Actually, when the downlink data is transmitted to the slot and the hardware is instructed, the physical slot / virtual slot comparison table 111 is referred to and an instruction is issued to the slot of the appropriate system.

  The instruction is sent to the slot control module 107 of each system, and a command is sent to the actual package of each system.

  As described above, by introducing modules that control and manage slots on the network, it is not necessary to be aware of the network for most of the processing of the main device, as if it controls your own system. It becomes possible to control the hardware.

  Note that the number of virtual slots is not limited by hardware as in the case of physical slots, and can be assigned indefinitely as long as the system memory permits.

  In addition, in the system, processing is usually performed with a virtual slot number. However, there are cases where it is desired to identify which slot of which system in a portion visible to the user, such as setting of system data.

  In that case, setting or the like can be performed in the physical slot with reference to the physical slot / virtual slot comparison table 111.

[Embodiment 2]
In the first embodiment, in order to avoid the problems of the distributed networking system, many of the problems that have been hitherto have been solved by inventing the centralized networking system.

  At the same time, however, it has the disadvantage of system vulnerability. That is, in the first embodiment, due to its nature, a master device that is a master (hereinafter referred to as “master main device”) centrally manages all resources on the network, and a master device that is a slave (all master devices). (Hereinafter referred to as “slave master device”), if the master master device fails due to some failure, all slave master devices connected to it become inoperable and the entire network Had the problem of becoming.

  The second embodiment is an improvement of the first embodiment, and aims to increase the robustness of the entire system in preparation for a failure such as a network malfunction by providing the networking system with redundancy. Is.

  In the second embodiment, when a master device serving as a master main device fails, a proxy main device is selected from slave main devices connected to the master device, and the main device operates as a master main device. This avoids the above problem.

  In other words, if the master master unit fails, the slave main unit connected to it detects it and selects a proxy main unit from all slave main units according to the preset priority. To do. Then, the slave main device changed to the proxy master main device starts operation as the master main device and controls all the slave main devices. By introducing this mechanism, even if the master main unit malfunctions, the entire network will not malfunction, and it will be possible to continue operation and enhance the robustness of the system.

  In the second embodiment, in the centralized management networking, when the master main unit system malfunctions, the proxy master main unit is uniquely selected and selected based on the priority set for each main unit. Take over the master main unit function to the main unit.

  9 and 10 are conceptual diagrams of the present invention.

  Each arrow indicates a communication relationship, and bidirectional communication is always performed between the master main unit / slave main unit.

  When communication is interrupted due to a malfunction of the master main device or a network failure or the like, the master main device is replaced and changes from the state shown in FIG. 9 to the state shown in FIG.

  The ellipses in FIGS. 9 and 10 represent main devices, and master main device priorities are set as system data in each main device. In the event of a failure, the master device with the highest master main device priority is selected as the master main device from the operating main devices and operates. Further, mutual IP addresses are set for all the main apparatuses.

  Each slave main unit constantly monitors the connection with the master main unit. If the master main unit cannot communicate with the master main unit for a certain period of time, the master main unit malfunctions or a network failure occurs. It is considered that it has been selected, and it will be elected as the master master device of the substitute.

  The selection of the substitute master main device is performed by each main device inquiring the priority order of all the main devices other than itself. The inquiry is made by throwing a specific packet on the IP to the list of other main devices (represented by IP addresses) of each main device.

  The main apparatus that has received this packet returns its priority.

  FIG. 11 shows a conceptual diagram of the priority order inquiry.

  Each ellipse represents a main apparatus that is inquiring about priority, and an arrow represents that an inquiry is being made. As shown in the figure, the priority order is inquired by the brute force method.

  Responses are returned from all the main devices in the list that the user has, and if the device itself has the highest priority, the device starts operating as the master main device. The main device that does not return a response after a certain period of time is considered to be malfunctioning and is judged among those that have returned a reply.

  If the device itself becomes the master main device, thereafter, in response to inquiries, it notifies that it is the master main device.

  On the other hand, the main device having the lowest priority keeps inquiring until the main device having the highest priority becomes the master main device. While the inquiry continues, the main device that becomes the master main device starts to operate, and in response to the inquiry, the new master main device returns a reply that it is the master main device. When this reply is received, the master apparatus is connected to the master apparatus that sent the reply, and the main apparatus starts operating as a slave main apparatus.

  The above flow is continued until oneself becomes a master master device or a slave main device. By repeating this process, the master master device with the highest priority is uniquely selected, and the selected master master device is the center. New networks can be formed.

  Each main apparatus is connected by an IP network, and has a unique IP address and a master main apparatus candidate priority value. It also has a list of IP addresses of all the other main devices that make up the network. Further, in order to identify each main device, each main device has a unique ID. Hereinafter, this ID is referred to as a system ID. The system ID may be considered equal to the master main device priority.

  FIG. 12 shows a configuration diagram of the present invention.

  As shown in FIG. 12, each main apparatus is uniquely assigned an IP address and a system ID, and has information on the IP address and system ID of another main apparatus as data.

  When the master main device has not yet been determined, such as when the master main device has malfunctioned or at the time of network construction, the master main device is determined based on these pieces of information.

  With this method, it is possible to select a proxy master main device and continue to operate as a network as a minimum, but this method alone is not sufficient to operate without impairing the function. Next, a method for compensating for this will be described.

1. System data synchronization The main unit holds setting information such as telephones and lines as files called system data.

  In centralized networking, the master main device manages and operates information on all slave main devices, so the master main device refers to system data held by the master main device.

  Therefore, when the master main unit is switched, the main unit that became the new master main unit refers to the system data of its own main unit, so the main unit that can be the master main unit is always operating as the current master main unit. It is necessary to keep the same system data as the operating system.

  In order to realize this, in the centralized management type networking, every time the master main unit updates the system data, a mechanism is introduced that always transmits and synchronizes the latest data to the slave main unit.

2. Forced master main unit switching In the master main unit replacement method described above, if the master main unit cannot be connected for a certain period of time, it is considered that the master main unit has failed, and the master main unit is switched automatically. In some cases, the master main device may be reset to the original main device.

  In addition, when the network system is initially constructed, which main device becomes the master main device varies depending on the order of startup, which is inconvenient when it is desired to make a specific main device the master main device.

  For such a case, a function for forcibly setting the master main unit is provided.

  The master main device is forcibly set by manually specifying the system ID of the main device desired to be the master main device by the user. For example, when a certain number (such as # 999) + system ID is pressed on a certain telephone terminal, the main device having that system ID is forcibly set as the master main device.

  This operation can be performed from a terminal of any system connected to the network.

  When this operation is performed, the master main unit can determine which system ID is the new master main unit, so it acquires the IP address of the new master main unit from its own IP address list, and executes the command A command to the effect that the transmission destination is the master main device is transmitted as a packet.

  The main device that has received the command starts operation as the master main device, and notifies all the main devices that the master main device has been changed.

  According to the second embodiment, it is possible to avoid the problem of vulnerability that the entire network becomes malfunctioning when the master main apparatus malfunctions, which has been held in the network of the central management system.

  This is a particularly important function in a system such as a telephone that can be used for mission-critical tasks, and can be said to be indispensable.

[Embodiment 3]
In the second embodiment, by providing a redundancy function to the network of the resource centralized management system, the master main device that controls the entire network can be switched, and the robustness of the system is improved.

  However, in the second embodiment, at the same time, when the master main unit is changed, communication terminals such as an external CTI (Computer Telephony Integration) server and ACD-MIS (Automatic Call Distributor-Management Information System) are connected to each other. The other party changes, and you have the problem that you can no longer communicate. Further, communication terminals such as an external CTI server and ACD-MIS need to know in advance which main device of the network is the master main device.

  In the third embodiment, each main device other than the master main device transfers (redirects) communication from a communication terminal such as a CTI server or ACD-MIS to the master main device to avoid the above problem.

  That is, the main device that recognizes the connection from the communication terminal is, if it is not the master main device, that is, if it is a slave main device, communicates with the master main device to which it is connected Redirect connections from. Also, the data from the communication terminal is transferred, and the data from the master main unit is relayed to the communication terminal.

  By doing so, it looks as if the communication terminal is communicating with the master main device, and the master main device appears to be communicating with the communication terminal.

  As a result, the communication terminal need not be aware that the master main device has been switched, and need not be aware of which main device is the master main device on the network.

  And as a master main apparatus, since it seems to be connected with the communication terminal as before, there is no need to change the process.

  Hereinafter, Embodiment 3 will be described in more detail with reference to the drawings.

  FIG. 13 shows a conceptual diagram of connection with a communication terminal in networking of the resource centralized management method.

  There is only one master main device that performs all call control and resource management on the network, and the main device that is the slave main device communicates with the master main device and follows all instructions from the master main device.

  The communication terminal communicates with the master main apparatus according to a preset IP address, acquires call information, and controls the main apparatus.

  Here, when communication is interrupted or the system is down, the master main unit is replaced by the redundancy function. A conceptual diagram of this state is shown in FIG.

  At this time, the main device that was the slave main device becomes the master main device, and the main device that was the master main device becomes the slave main device, but the communication terminal is normally only configured to communicate with a single main device. Therefore, even if the master main unit is switched, there is no choice but to communicate with the other party as usual.

  However, since the communication partner is not a master main unit but a slave main unit that only follows the instruction from the master main unit, it can communicate with the communication terminal, but cannot return appropriate information.

  Therefore, in the present embodiment, a technique is provided in which the connection of the communication terminal is relayed so that the communication terminal looks like a new master main device and the new master main device communicates directly with the communication terminal.

  In the configuration diagram shown in FIG. 2, each main device is connected by the Internet protocol, and the master main device communicates with each slave main device. The master main unit manages all system information such as call information and resource information. However, the slave main unit basically has only a communication function, and performs appropriate processing directly on the communication terminal. Can not do.

  In addition, the communication terminal can only connect to a specific main device.

  The actual operation of the third embodiment will be described.

  FIG. 15 shows a conceptual diagram of communication between a conventional communication terminal and a main apparatus.

  The main device opens a specific port set by system data for connection of the communication terminal, and waits for connection at the port.

  The communication terminal connects to the IP address and port of the main device.

  This is performed by the application of the communication terminal, and the IP address and port need to be set in advance.

  For example, when performing CTI, the main apparatus prepares a communication port 8000 for CTI and waits. On the other hand, the communication terminal starts up the CTI application, sets the IP address of the main apparatus and the communication port 8000, and connects to the main apparatus.

  Then, a command is transmitted to the main apparatus by transmitting data, and a result is obtained by receiving data.

  On the other hand, the conceptual diagram of Embodiment 3 is shown in FIG.

  When the main apparatus receives a connection from the communication terminal, if it is not the master main apparatus, it connects to the master main apparatus while maintaining the connection from the communication terminal.

  At this time, since the system data is synchronized between the master main unit (former slave unit) and the slave main unit (previous master unit), the slave main unit uses the same port as the master main unit from the communication terminal. It can wait for a connection and can connect to a standby port of the master main unit.

  The master main device that has received the connection from the slave main device recognizes that the connection is from the communication terminal, and starts the service. On the other hand, the communication terminal side recognizes that it is connected to the master main unit by connecting to the slave main unit, and starts operation.

  Thereafter, the slave main device transfers data from the communication terminal to the master main device, and transfers data from the master main device to the communication terminal. This makes it possible for the communication terminal to communicate without being aware of it wherever the master main unit is actually located.

  Therefore, even if the application of the communication terminal is made by a third party, it can be used without any problem. Further, since the master main apparatus recognizes the communication terminal as if it was connected as before, there is no need to change the process.

  In addition, as shown in FIG. 17, the communication terminal can communicate with the master main apparatus regardless of which main apparatus is connected, so the communication terminal registers any one main apparatus in the network as a connection destination. You don't have to be aware of the master master device.

  By utilizing the present invention, it is possible to avoid the problem that the communication terminal cannot be connected due to the master main device replacement by the redundancy function. In addition, since the communication terminal and the master main device can be assured to operate as before, there is no need to change the application.

  According to the third embodiment, it is possible to avoid the problem of vulnerability that the entire network becomes malfunctioning when the master main apparatus malfunctions, which has been held in the network of the central management system.

  This is a particularly important function in a system such as a telephone that can be used for mission-critical tasks, and can be said to be indispensable.

[Embodiment 4]
The operation when the master main unit goes down has been described in the second and third embodiments. On the other hand, the fourth embodiment is for the case where the system is restored.

  In the fourth embodiment, in the network state temporarily divided due to a failure or the like, the network is integrated by automatically detecting the recovery, and the state is automatically shifted to the original operation state.

  In the second embodiment and the third embodiment, when a master main device in networking without function restriction by centralized resource management (hereinafter referred to as centralized management networking) goes down or communication becomes impossible due to a network failure, the master master acting as a proxy By selecting a device from among the slave main devices, we were able to continue operation without bringing down the entire network.

  On the other hand, if a master master device is selected as a proxy, the proxy master main device continues to operate as a proxy master main device even if communication between the original master main device and another main device is restored. Therefore, although two or more master main devices can be created and the operation can be continued, the communication between the two main devices can be performed, and the network remains divided.

  In the fourth embodiment, after a network failure occurs, when the communication between the proxy master main device and the original master main device is restored, it is automatically detected, and the network is reconstructed around the original master. Avoid the above problems.

  In other words, the proxy master master device periodically monitors the status of the original master master device, and when the recovery of the original master master device is detected, the proxy master master device itself and the slave master device under its control Reconnect the original master master as a slave master. By doing so, the network is automatically integrated after the communication is restored, and the operation can be continued in the same state as before the failure.

  18, 19 and 20 show conceptual diagrams of the fourth embodiment.

  As shown in FIG. 18, in the centralized management networking, the master device as a master centrally manages and controls the resources of the master device as a slave. For this reason, when the master main unit goes down, there is no main unit that performs control, and therefore, as shown in the second and third embodiments, the proxy master is selected and the operation is continued. The mechanism was invented (FIG. 19). In the fourth embodiment, as shown in FIG. 20, when the original master main apparatus is restored, a mechanism capable of reconstructing the network and continuing the operation as originally described will be described in detail.

  FIG. 21 shows a conceptual diagram. When Fail-Over (selection of alternative master main device at the time of failure and network reconfiguration) occurs and a network centering on the proxy master main device is formed, the proxy master main device An IP address of the master main apparatus is acquired, and a function for monitoring the return of the original master main apparatus (hereinafter referred to as “the highest priority master main apparatus”) is executed.

  As for the IP address of the highest priority master, if the IP address of the highest priority master itself is set in advance, the data is automatically transmitted to each node (main device) and synchronized, so that Fail-Over is taken. Then, the return of the highest priority master device can be monitored based on the IP address.

  FIG. 22 shows a conceptual diagram of a task configuration for monitoring master return.

  When the main device that has become the proxy master main device starts operation as the proxy master main device, it compares its own IP address with the IP address of the highest priority master, and if it does not match, it is itself the proxy master main device And the task for monitoring the return of the highest priority master device is started. This task runs periodically at a low priority and keeps trying a single task that tries to connect to the IP address of the highest priority master.

  During this time, tasks that control the slave main unit as the master main unit operate normally, and while monitoring the return of the highest priority master main unit, control the slave main unit as the master main unit and continue to operate networking. be able to.

  FIG. 23 shows a conceptual diagram of the operation when the highest priority master main device recovers and the proxy master main device detects it.

  When the recovery of the highest-priority master main unit is detected by the highest-priority master main unit monitoring task, the substitute master main unit sets itself as the slave main unit, and the master main unit that controls the slave main unit becomes the proxy master unit. The operation for changing from the master device to the highest priority master device is started.

  As shown in FIG. 23, first of all, it changes itself from the master main unit to the slave main unit and reconnects to the highest priority master main unit.

  At this time, the highest priority master main device monitoring task and the call control task are no longer necessary and are deleted, and the slave control task is switched to the master communication task. Then, a hardware control task is generated, and the operation mode is switched so that the device itself operates as a slave main device. Then, it notifies the slave that is under its control of the IP address of the highest priority master main unit, instructs the slave to switch to the highest priority master main unit, and itself also sets the slave to the highest priority master main unit. Connect to the main unit.

  The slave main device that has received the instruction reconnects to the highest priority master main device from the substitute master main device, and switches the operation so that it is controlled by the highest priority master main device.

  By the above operation, the proxy master main device and the slave main devices under the proxy master main device can reconnect to the highest priority master main device and return to the connection state before Fail-Over. As a result, the operating state can be quickly recovered after the network is restored.

  Here, a network configuration as shown in FIG. 24 is considered.

  Black dots indicate communication devices such as routers that interconnect networks.

  When a communication failure occurs between networks as shown in FIG. 25, the second and third embodiments present a method for reconfiguring a network so as to configure the maximum network between communicable nodes. As a result, when a communication failure occurs between the networks indicated by the crosses in the figure, the networks as shown by the two ellipses in the figure are respectively configured, and each master main unit is limited. Could operate as a network.

  According to the fourth embodiment, when a failure between networks is recovered, the lower isolated network can detect the recovery of the upper master device by using the method shown in FIGS. The connection state shown in FIG. 24 can be restored. As a result, while operating on a limited network in the event of a failure, a complete network connection can be restored when the failure is recovered.

  In FIG. 24, black dots represent connection devices that interconnect networks such as routers. Different networks are interconnected by routers, and the master master device centrally manages the information of all nodes to realize networking. Therefore, when a failure occurs in communication between networks, the Fail-Over function operates so as to take the maximum configuration in the same network as shown in FIG.

  In the network configuration shown in FIG. 24, when a failure occurs in communication between networks, the fail-over function allows the proxy master to perform maximum communication in the divided network as shown in FIG. The main unit is selected and the network is reconfigured. The main device that has become the proxy master main device starts the monitoring operation of the highest priority master main device shown in FIG. 22 from the time when the operation starts as the proxy master main device, and starts monitoring the recovery of the highest priority master.

  When the highest-priority master main device or the network is restored, a response from the highest-priority master main device can be received, so that failure recovery is determined and operation for recovery is started.

  The slave master unit operating under the subordinate master master unit is replaced by the master master unit that receives control. Notify the IP address of the new master main unit.

  The slave main unit basically operates as a slave main unit as before, and only the connection destination changes, so no special processing is required at this time. However, when the master main unit becomes a slave main unit, it is necessary to stop various tasks to operate as a proxy master unit, shift to the slave task, and release various resources that were centrally managed as the master main unit. It is. Here, the resource release will be described.

  As shown in FIG. 5, in the centralized management networking, a method is used in which the resources of the nodes connected to the network are centrally managed by the main device as the master main device.

  Therefore, when the role of the master is finished, the management state of these resources is cleared.

  After these tasks and resources are cleared, the proxy master main device shifts to the slave main device, reconnects to the highest priority master main device, and receives control. Since the state as the proxy master main device is internally cleared, it is not necessary to physically reset the system, and a quick recovery operation is possible.

  Through the above processing, it is possible to quickly recover completely to the state before the failure and continue the operation.

  By applying this embodiment, it becomes possible to automatically detect the failure recovery and return to the original operation state. In addition, failure recovery can always be monitored and recovery processing can be performed quickly, so it is possible to construct a system that always operates in an optimal state.

  The recovery process does not require a physical reset of the system, but it is necessary to reset the package that controls the terminal / line when the task is switched or the connection destination is changed. It is not possible. Therefore, it is also necessary to perform the recovery process manually when the system is in an idle state, not automatically. Another possible method is to monitor the system status, detect recovery, and perform recovery processing when the system becomes idle.

[Embodiment 5]
In the first embodiment, in order to avoid the problem of the distributed networking system, the centralized management type networking method is used as a reference form, thereby solving many of the conventional problems.

  That is, by inventing networking of the resource centralized management system, information management becomes easy, and networking without function restrictions can be realized. However, the processing is concentrated on one main device, which increases the burden on the CPU and makes it impossible to control many resources. That is, all the systems on the network are controlled by one master main device, and a huge load is concentrated on the master main device.

  Therefore, the fifth embodiment realizes networking using a CPU on a personal computer (hereinafter referred to as “PC”) having high performance, reduces the load on the master main apparatus that requires a large amount of traffic, and external applications. The purpose is to improve the affinity.

  In the fifth embodiment, the above problem is avoided by using a high-performance general-purpose PC instead of an exchange as a system serving as a master main apparatus. That is, the same program as that of the exchange is executed on the PC, and each main apparatus and the PC are connected by the IP network. The hardware of the main device connected as the slave main device is controlled by the PC as the master main device.

  By doing so, it becomes possible to use the high-performance CPU of the PC and to process a huge amount of traffic from the slave main unit.

  FIG. 26 shows a conceptual diagram of networking in a resource centralized management system in which a PC is a master main device.

  FIG. 26 shows that the PC is operated as a master main device and the actual main device is controlled as a slave main device.

  Here, naturally, since the PC cannot directly control the hardware of the exchange, the control is performed by each connected slave main unit.

  This will be described with reference to FIG.

  The package management method of FIG. 27 applies the reference form.

  A package that is actual hardware is installed in system IDs 2, 3, and 4 that are slave main devices, but all of the information is sent to and managed by system ID 1 that is a master main device.

  In addition, all information such as terminals and lines connected to the package is also uploaded as upstream data to the master main unit.

  By doing so, it appears as if the package is installed on itself, even though it is not capable of accommodating the package, from the perspective of the master main unit.

  In addition, an instruction for an actual package is transferred to the package of each system, and the terminal, line, and the like are controlled.

  In this way, it is possible to control the exchange even with a general-purpose PC that cannot directly control the package, and because it can use a high-performance CPU, it can withstand traffic even when many slave main units are connected. Can do.

  This alleviates the problems of networking due to centralized resource management, and can build a larger-scale networking.

  In addition, in the networking of the resource centralized management system, when the master main device goes down for some reason or communication is interrupted, a mechanism for selecting a proxy master main device and continuing operation is introduced.

  In the mechanism of FIG. 26, when the PC that is the master main unit goes down, the PC is not connected to the network, so the master main unit candidate priority order set in advance among the main units connected as slave main units The master device with the highest CPU becomes the master main device, and as a result, the node with low CPU capability becomes the master main device.

  In order to prevent this, we will introduce a mechanism that puts the PC that is the proxy master main unit on standby and operates it when a failure occurs.

  FIG. 28 shows an embodiment.

  FIG. 28 shows that a PC is newly joined as a slave main device to the network shown in FIG.

  While the system 1 as the master main apparatus is operating normally, the connected PC with the system ID 5 is not operating and is on standby. When a failure occurs in the master main unit, it is detected, and the PC with the system ID 5 starts operating as a proxy master main unit. In this way, even if a failure occurs in the PC as the master main unit as shown in FIG. 29, a PC having the same CPU performance can take over the master main unit and reduce the traffic processing capacity. Can be prevented. When the system recovers from the failure, as shown in FIG. 30, the system 1 is connected to the system 5 as a slave main device and stands by in preparation for the failure. When a failure occurs in the system 5, the system ID1 PC is started again as the master main device.

  In this way, if either PC is operating, the network can be operated with the PC as the master main unit.

  The systems 2, 3, and 4 are actual main apparatuses having slots for installing packages, and accommodate packages that connect to terminals, public lines, and IP networks, respectively. These operate as slave main apparatuses, and are connected to a PC operating as a master main apparatus in the system 1 by IP and are controlled.

  The PC operating as the master main unit of the system 1 collectively manages hardware information such as the packages of the systems 2, 3, and 4 and information such as call status, and performs call control of all slave main units to be connected. I do.

  The system 5 is connected to the system 1 as a slave main device, and monitors and waits for the system 1 in case of a failure.

  Next, a specific implementation method will be described.

  A method related to package resource management is as described in the reference form.

  Here, other methods for using the PC as a master main apparatus will be described.

  FIG. 31 shows a conceptual diagram of networking when a conventional real main device is used as a master main device.

  The functional modules according to the present invention can be roughly divided into a call control unit that controls incoming calls and calls, a package that transmits and receives package data between a master main unit and a slave main unit, and data received from a communication unit and a package. Or a hardware control unit that gives instructions to the package.

  In FIG. 31, the call control unit is operating only in the main apparatus that is the master main apparatus, and the slave main apparatus is not performing call control.

  The slave main unit receives data from the package at the hardware control unit, and transmits data to the master main unit at the communication unit.

  The master main unit performs call control based on the data at the call control unit, loops back and transmits the package data at the communication unit, and the slave main unit receives the data and controls the hardware.

  Since the master main unit itself has a slot for accommodating the package, the hardware control unit operates and controls the package.

  FIG. 32 shows a conceptual diagram of functional modules in the present embodiment.

  Since the master main device is a general-purpose PC, naturally, the hardware control unit is unnecessary because it cannot directly accommodate the package.

  In addition to being unnecessary, the operation itself must be prohibited in order to control hardware that does not exist.

  Therefore, since the modules to be mounted are only the call control unit and the communication unit as shown in the figure, they can be mounted on a PC without depending on the processing target.

  Further, since communication is performed between the master main apparatus and the slave main apparatus using TCP / IP, the OS does not have to be the same between the master main apparatus and the slave main apparatus as long as they are equivalent in terms of processing. For example, it is possible to install an embedded OS such as Linux (registered trademark) on a PC and VxWorks (registered trademark) or Nucleus Plus on an actual main apparatus.

  As shown in the figure, the main device serving as the slave main device basically operates only the communication unit and the hardware control unit.

  Therefore, a high-performance CPU is required for the master main unit PC, but an inexpensive CPU is sufficient for the main unit serving as the slave main unit. Similarly, since all resource management is concentrated on the master main unit, the PC that becomes the master main unit needs many memory resources, but the main unit that becomes the slave main unit does not perform any resource management. It is possible to operate with very few memory resources.

  In addition, in the networking of the resource centralized management method, there is a function to select and operate a proxy master main device from the master master device with a high priority set in advance when the master main device goes down, but it is poor. It is necessary to assign a low priority to the real main device and assign a high priority to the backup PC so that the real main device having only the CPU and memory is not elected as the proxy master main device.

  By applying the fifth embodiment, a system having a high-performance CPU can be used as a master main device, and a larger-scale networking can be constructed.

  Also, since the main device software can be executed on the PC, the affinity between the PC and the main device is increased, and application cooperation such as CTI (Computer Telephony Integration) becomes easy.

[Embodiment 6]
In the sixth embodiment, a more specific configuration will be described.

  FIG. 33 shows a group of main devices according to a conventional example. Referring to FIG. 33, according to the conventional example, main devices included in the main device group are not classified into master main devices and slave main devices. Each main device operates independently and is connected by an IP network 203 via a VoIP interface 201. Each main apparatus includes a CAPS / OPMS 105, an IOCS 103, and a slot I / F 101, but does not include a slot management module 109, a physical slot / virtual slot comparison table 111, and a slot control module 107.

  FIG. 34 shows a main apparatus group according to an embodiment of the present invention. Referring to FIG. 34, according to the embodiment of the present invention, the main devices included in the main device group are classified into one master main device 211 and a plurality of slave main devices 213. One master main apparatus 211 and a plurality of slave main apparatuses 213 are connected by an IP network 217 via a VoIP interface 215. The plurality of slave main devices 213 operate under the control of one master main device 211. The master main unit 211 includes a CAPS / OPMS 105, an IOCS 103, a slot management module 109, a physical slot / virtual slot comparison table 111, and a slot I / F module 101. Each slave main device 213 includes a slot control module 107 and a slot I / F 101.

  FIG. 35 shows an example of the appearance of the main device. The main device has a plurality of physical slots, and packages are inserted into the physical slots.

  FIG. 36 shows an example of the appearance of a package. The package includes a connector 303 connected to the backboard connector 301 (see FIG. 35) of the physical slot, and a connector 305 connected to an external line or a PBX internal line.

It is a figure which shows a mode that the real package of another main apparatus is handled as a real package of a self-device using the virtual package by embodiment of this invention. It is a figure which shows the example of a connection of the master main apparatus and slave main apparatus by embodiment of this invention. It is a figure which shows the other example of a connection of the master main unit by the embodiment of this invention, and a slave main unit. It is a figure which shows the correspondence of the virtual slot and physical slot by embodiment of this invention. It is a figure which shows the example of the correspondence of a virtual slot and a physical slot by the embodiment of this invention, and the connection destination of each physical slot. It is a figure which shows the connection relation with CAPS / OPMS, IOCS, and a slot interface by a prior art example. It is a figure which shows the connection relationship with CAPS / OPMS, IOCS, a slot management module, a physical slot / virtual slot comparison table, a slot control module, and a slot interface by embodiment of this invention. It is a figure which shows the specific example of the physical slot / virtual slot comparison table by embodiment of this invention. It is a 1st figure which shows the connection relation of the master main apparatus and slave main apparatus by Embodiment 2 of this invention. It is a 2nd figure which shows the connection relation of the master main apparatus and slave main apparatus by Embodiment 2 of this invention. It is a figure which shows the mode of the inquiry of the priority between the main apparatuses by Embodiment 2 of this invention. It is a 3rd figure which shows the connection relation of the master main apparatus and slave main apparatus by Embodiment 2 of this invention. It is a conceptual diagram of the connection with the communication terminal in networking of a resource centralized management system. It is a figure which shows the connection relation of the communication terminal, master main apparatus, and slave main apparatus when the master main apparatus switches by embodiment of this invention. It is a conceptual diagram of communication between the conventional communication terminal and the main apparatus. It is a conceptual diagram of communication between the communication terminal and main apparatus by Embodiment 3 of this invention. It is a figure which shows that the communication terminal by Embodiment 3 of this invention can be connected to any main apparatus. It is a figure which shows the connection relation between the main apparatuses at the normal time in Embodiment 4 of this invention. It is a figure which shows the connection relation between the main apparatuses when a failure generate | occur | produces in the highest priority master main apparatus in Embodiment 4 of this invention. It is a figure which shows the connection relation between the main apparatuses when the highest priority master main apparatus in Embodiment 4 of this invention recovers from a failure. In Embodiment 4 of this invention, it is a figure which shows that the substitute master main apparatus monitors the return | restoration of the highest priority master main apparatus in which a failure generate | occur | produced. It is a conceptual diagram about the task structure which monitors master return by Embodiment 4 of this invention. It is a conceptual diagram about operation | movement when the highest priority master main apparatus recovers and it is detected by the proxy master main apparatus by Embodiment 4 of this invention. It is a figure which shows the example of the network by which the master main apparatus and slave main apparatus at the time of normal by Embodiment 4 of this invention were connected. It is a figure which shows the example of the network with which the master main apparatus and the slave main apparatus were connected when the part of network by Embodiment 4 of this invention was cut | disconnected. It is a conceptual diagram of the network of the resource centralized management system which used PC by Embodiment 5 of this invention as a master main apparatus. It is a figure which shows the package management system by Embodiment 5 of this invention. FIG. 9 is a conceptual diagram of networking in a resource centralized management system in which a first PC according to Embodiment 5 of the present invention is a master main device and a second PC is a proxy master device. In the form shown in FIG. 28, it is a figure which shows the connection relation between the main apparatuses when 1st PC becomes malfunctioning. FIG. 30 is a diagram illustrating a connection relationship between main apparatuses when the first PC is restored in the state illustrated in FIG. 29. It is a conceptual diagram of networking when the conventional real main device is used as a master main device. FIG. 3 is a conceptual diagram of networking when a PC is a master main device. It is a block diagram which shows the main apparatus group by a prior art example. It is a block diagram which shows the main apparatus group by embodiment of this invention. It is a perspective view which shows the external appearance of the main apparatus by embodiment of this invention. It is the front view and side view which show the external appearance of the package by embodiment of this invention.

Explanation of symbols

101 System interface 103 IOCS (input / output control module)
105 CAPS (call control module) / OPMS (package, terminal management module)
107 slot control module 109 slot management module 111 physical slot / management slot comparison table

Claims (22)

A slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and the lower slot interface.
The input / output control module accesses the slot interface using virtual slot identification information;
The slot management module refers to the physical slot / management slot comparison table, converts virtual slot identification information into physical slot identification information, and accesses a slot control module corresponding to the physical slot identification information obtained by the conversion. Thus, physical access to the slot interface by the input / output control module is realized,
When a new slot interface is added, a new virtual slot is added, and the correspondence relationship between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is set as the physical slot. A slot interface access apparatus, further comprising means for adding to a slot / management slot comparison table. The slot interface access device according to claim 1,
The slot interface access device, wherein the slot interface and the slot control module are distributed in a plurality of main devices. A slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the I / O control module and the lower slot interface.
The input / output control module accesses the slot interface using virtual slot identification information;
The slot management module refers to the physical slot / management slot comparison table, converts virtual slot identification information into physical slot identification information, and accesses a slot control module corresponding to the physical slot identification information obtained by the conversion. Thus, physical access to the slot interface by the input / output control module is realized,
When a new slot interface is added, a new virtual slot is added, and the correspondence relationship between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is set as the physical slot. A slot interface access method, further comprising the step of adding to a slot / management slot comparison table. The slot interface access method according to claim 3,
The slot interface access method, wherein the slot interface and the slot control module are distributed in a plurality of main devices.   A program for causing a computer to function as the slot interface access device according to claim 1. A plurality of main devices having the slot interface access device according to claim 1 or 2, wherein one main device is a master main device and the other main device is a slave main device,
If the master master device fails, one slave master device of the slave master devices becomes the new master master device,
Each main unit has a priority, and if the current master main unit fails, the main unit with the highest priority among the other main units shall be the new master main unit,
Each main unit transmits a priority inquiry signal to all other main units in a brute force manner, compares the priority obtained by replying to the signal with the priority of the main unit itself, When the priority is higher than the priority of all other main devices, the main device itself becomes the master main device, and in response to an inquiry from another main device, the main device itself notifies the master main device. ,
Each main unit recognizes that the main unit itself is a slave main unit when receiving a notification from the new main unit that the main unit is the master main unit. . In the redundant configuration of the main device according to claim 6,
A redundant configuration of the main unit characterized in that the system data held by the master main unit is also held by the slave main unit which is another main unit. In the redundant configuration of the main device according to claim 6 or 7,
A redundant configuration of a main device, wherein communication between a terminal and a master main device is routed through a slave main device. In the redundant configuration of the main device according to any one of claims 6 to 8,
The new master master device monitors whether the master master device that has failed has recovered,
A redundant configuration of the main unit characterized in that when the master master unit that has failed is recovered, the new master main unit returns control of the entire system to the recovered master main unit. In the redundant configuration of the main device according to claim 9,
The new master master unit queries the master master unit that has become dysfunctional and recovers the master master unit that has failed if a response is obtained from the master unit that has failed. A redundant configuration of the main device, characterized in that A plurality of main devices having the slot interface access device according to claim 1 or 2, wherein one main device is a master main device and the other main device is a slave main device,
If the master master device fails, one slave master device of the slave master devices becomes the new master master device,
Each main unit has a priority, and if the current master main unit fails, the main unit with the highest priority among the other main units shall be the new master main unit,
Each main unit transmits a priority inquiry signal to all other main units in a brute force manner, compares the priority obtained by replying to the signal with the priority of the main unit itself, When the priority is higher than the priority of all other main devices, the main device itself becomes the master main device, and in response to an inquiry from another main device, the main device itself notifies the master main device. ,
Each main unit recognizes that the main unit itself is a slave main unit when receiving a notification from the new main unit that the main unit is the master main unit. . An alternative method of the main device according to claim 11,
An alternative method of a main device, characterized in that system data held by a master main device is also held by a slave main device which is another main device. An alternative method of the main device according to claim 11 or 12,
An alternative method of a main device, characterized in that communication between a terminal and a master main device passes through a slave device. An alternative method of the main device according to any one of claims 11 to 13,
The new master master device monitors whether the master master device that has failed has recovered,
An alternative method of the main device, wherein when the master master device that has failed is recovered, the new master main device returns the control right of the entire system to the recovered master main device. 15. An alternative method of the main device according to claim 14,
The new master master unit queries the master master unit that has become dysfunctional and recovers the master master unit that has failed if a response is obtained from the master unit that has failed. An alternative method of the main device, characterized in that it is determined that A slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and the lower slot interface.
The input / output control module accesses the slot interface using virtual slot identification information;
The slot management module refers to the physical slot / management slot comparison table, converts virtual slot identification information into physical slot identification information, and accesses a slot control module corresponding to the physical slot identification information obtained by the conversion. Thereby, physical access to the slot interface by the input / output control module is realized,
When a new slot interface is added, a new virtual slot is added, and the correspondence relationship between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is set as the physical slot. And further comprising means for adding to the slot / management slot comparison table;
A slot interface access device characterized in that the device has a higher CPU capacity than other devices having a slot interface. The slot interface access device according to claim 16,
The apparatus does not include the slot interface, and has a slot interface access apparatus.   18. The slot interface access device according to claim 16 or 17, and a slot interface access device that acts as a substitute for the slot interface access device when the slot interface access device malfunctions. A slot interface access device group comprising: a device having a CPU capability higher than that of the device. A slot management module, a slot control module, and a physical slot / management slot comparison table are provided between the input / output control module and the lower slot interface.
The input / output control module accesses the slot interface using virtual slot identification information;
The slot management module refers to the physical slot / management slot comparison table, converts virtual slot identification information into physical slot identification information, and accesses a slot control module corresponding to the physical slot identification information obtained by the conversion. Thereby, physical access to the slot interface by the input / output control module is realized,
When a new slot interface is added, a new virtual slot is added, and the correspondence relationship between the identification information of the slot control module corresponding to the added slot interface and the identification information of the added virtual slot is set as the physical slot. And further comprising a procedure for adding to the slot / management slot comparison table;
A slot interface access method, wherein the method is performed by a device having a higher CPU capacity than other devices having a slot interface. The slot interface access method according to claim 19,
An apparatus for performing the method does not include the slot interface.   21. When a device performing the slot interface access method according to claim 19 or 20 malfunctions, a device having a CPU capability higher than that of another device performs the method on behalf of the device. Slot interface access method.   A program for causing a computer to function as the slot interface access device according to claim 16 or 17.

Images