JP4617203B2 - Server apparatus and communication connection method - Google Patents

Description

  The present invention relates to a server device and a communication connection method. The server device of the present invention particularly relates to a distributed system of SIP (Session Initiation Protocol) servlet engine. The communication connection method of the present invention particularly relates to a multihome network address setting method in the SIP servlet engine.

  The SIP servlet engine accommodates one or two subnetworks. A SIP location server is connected to the SIP servlet engine. A plurality of user terminal devices are connected to each subnetwork. Of the connected user terminal devices, a procedure for starting a call between the user terminal devices that are logging in will be briefly described. The transmission source user terminal device sets and transmits SIP-URI (Uniform Resource Identifiers) as corresponding destination information to the communication partner, that is, the destination user terminal device, and transmits the subnetwork, one SIP servlet engine and The message is sent to the destination user terminal device via the subnetwork. At this time, the SIP servlet engine transmits an IP (Internet Protocol) address set to itself to the SIP location server, and sets this IP address in the SIP location server.

  The SIP servlet engine requests an actual destination from the SIP-URI as the destination information supplied to the SIP location server as SIP request processing. In the SIP location server, the destination information corresponding to the SIP-URI according to the request is obtained by reference. In response to the SIP request, the SIP location server outputs not only the destination information of the reference result but also user information such as the login state of the destination and the IP address being logged in to the SIP servlet engine.

  The SIP servlet engine calls the destination user terminal device based on the user information obtained corresponding to the destination SIP-URI. The SIP servlet engine outputs an error to the transmission source user terminal device when the destination user terminal device is not registered or is not logged in. The destination user terminal device connects in response to a call from the transmission source user terminal device and starts a call.

  Further, it may be a distributed system having a plurality of SIP servlet engines. In this case, a SIP application server is connected to the subnetwork. The SIP application server includes a SIP load balancer and a plurality of subnets. The SIP load balancer has a plurality of logical IP addresses, and has a function of distributing SIP requests to SIP servlet engines included in each subnet in accordance with load balancing processing rules. A plurality of SIP servlet engines are connected to each subnet, and a SIP location server is connected to each SIP servlet engine. The SIP servlet engine handles address port management information.

  The address port management information includes a plurality of transport information and load balancer information. The transport information includes an IP address, a port number, a protocol type, and an activation state. The IP address is the IP address of the SIP servlet engine network interface card. The port number is a number to be opened. The protocol type indicates whether the protocol is TCP (Transmission Control Protocol) / UDP (User Datagram Protocol). The activation state is represented by either “valid” or “invalid”, and indicates a busy state and an unused state, respectively.

  The load balancer information includes a load balancer address, a port number of each protocol, and an activation state. In the load balancer address, the logical IP address of the load balancer is set. The port number of each protocol is, for example, a port number opened by TCP and a port number opened by UDP. In addition, the activation state indicates the state by “valid” and “invalid” similarly to the transport information.

  When installing multiple SIP servlet engines in a subnet, a SIP load balancer is almost always installed. This installation shortens the service stop time when a failure occurs.

  The operation in this distributed system will also be briefly described. Also in this case, the transmission source user terminal device designates the SIP-URI to the destination user terminal device and transmits the call. The transmitted SIP-URI is supplied to the SIP load balancer via a subnet between the SIP application servers. The SIP load balancer designates the logical IP address of the SIP load balancer in the SIP location server, and sends a SIP request process to the SIP servlet engine corresponding to this designation. SIP request processing is performed by distributing to each corresponding SIP servlet.

  The designated SIP servlet obtains the actual destination information from the SIP-URI of the destination and accesses the SIP location server in the subnet. This SIP location server processes the SIP request with reference to the destination information. The SIP location server returns response information for the SIP request toward the requested logical IP address.

  Upon receiving the response, the SIP servlet engine extracts the destination user information from the response information. The SIP servlet engine calls the destination user terminal device based on the obtained user information. The SIP servlet engine outputs an error to the transmission source user terminal device when the destination user terminal device is not registered or is not logged in. The destination user terminal device connects in response to a call from the transmission source user terminal device and starts a call. As a result, the SIP servlet engine enables load distribution and service continuation in the event of a failure of the SIP application server.

  Patent document 1 and patent document 2 are mentioned as the address setting of a network. The network address setting method of Patent Document 1 discloses a configuration for a case where a routing node and sub-networks under the routing node are moved together or for address management in a multihomed format. The network address setting method includes an address server and a routing node, and the address server assigns a physical address in the network and an address of the subnetwork to the routing node. The routing node sets the address assigned by the address server in the subnetwork. As the sub-network moves, the address server gives the destination physical address to the moved routing node.

  The multi-home service system of Patent Document 2 includes an application processing unit, a network processing unit, and a main processing unit in the control unit, eliminates the risk of information leakage, and can also access devices connected to other home networks. Make it extensible.

Japanese Unexamined Patent Publication No. 2000-32028 JP 2004-320766 A

  By the way, in the distributed system, when accommodating user information with different network addresses, it cannot be accommodated in the existing SIP servlet engine. To accommodate different user information, a new SIP servlet engine should be provided. Thereby, it becomes possible to accommodate. However, in the construction of a new SIP servlet engine in this way, if the existing SIP servlet engine has available resources such as memory and CPU (Central Processing Unit), the available resources cannot be used effectively.

  It is an object of the present invention to provide a server apparatus and a communication connection method that can eliminate such drawbacks of the prior art, accommodate different network addresses in an existing SIP servlet engine, and efficiently use free resources of the engine. To do.

  In order to solve the above-described problems, the present invention provides a server apparatus having a database that provides information and distributed information processing function blocks connected to information apparatuses that operate according to a predetermined protocol. The functional block manages first information regarding a plurality of sub-networks to which the information processing functional block is connected, information management means for managing a request supplied based on this management and a response to the request, and first information Storage means provided with an identifier for associating both the second information and the third information with respect to the second information and the third information included in the information processing function. With reference to the availability of distribution means and the identifier of the third information, the presence or absence of the corresponding identifier and the availability of the activation status of the corresponding identifier are determined, Characterized by managing the supply of the response to the quest.

  The server device of the present invention refers to the identifier of the third information based on the identifier that associates both the second information and the third information with the second information and the third information included in the storage unit by the information management unit. Since it is possible to determine whether the corresponding identifier is in the activated state and the response to the request is known, information handled by different sub-networks can be accommodated in the information processing function block. When adding a sub-network, it is possible to accommodate new users by adding and setting the second and third information and identifiers to the existing information processing function block, and to effectively use existing resources. Can do.

  In order to solve the above-described problem, the present invention uses an information processing function block that operates in a predetermined protocol to which a plurality of sub-networks are connected, and a load distribution unit that distributes the load of the information processing function block. In the communication connection method for establishing the communication connection by returning the request supplied based on the first information managed by the information processing function block and the response to the request to the position having the corresponding address information of the load distribution means, The method sets a first identifier for associating both the second information and the third information with respect to the second information and the third information included in the first information, and determines whether the identifier is present. When there is an identifier, the process associates the third information corresponding to the second information via the identifier, and obtains address information included in the third information. The second step, the third step for determining whether or not the load distribution means included in the third information is in the activated state, and the address included in the third information if the load distribution means is in the activated state. And a fourth step of outputting information to the load distribution means.

  The communication connection method of the present invention sets an identifier, determines whether or not this identifier is present, and if there is an identifier, associates third information corresponding to the second information via the identifier, The address information included in the information of the third information is obtained, it is determined whether or not the load distribution means included in the third information is in an activated state, and if the load distribution means is in the activated state, the address information included in the third information is a response. By outputting to the load balancing means, it is possible to make a request-response to users of different sub-networks, and efficiently use information processing function block resources, that is, free resources. A new user can be accommodated by adding and setting the second and third information and the identifier in the information processing function block.

  Next, an embodiment of a server device according to the present invention will be described in detail with reference to the accompanying drawings.

  In the present embodiment, the server device of the present invention is applied to the SIP application server 12 of the multihomed network system 10. The illustration and description of parts not directly related to the present invention are omitted. In the following description, the signal is indicated by the reference number of the connecting line in which it appears.

  As shown in FIG. 1, the network system 10 includes user terminal devices 12, 14, 16, 18 and a SIP application server 20. The multihomed network system 10 in FIG. 1 is a distributed network system. The user terminal devices 12 and 14 are connected to the SIP application server 20 via the subnetwork 22. The user terminal devices 16 and 18 are connected to the SIP application server 20 via the subnetwork 24.

  The SIP application server 20 includes a SIP load balancer 26, SIP servlet engines 28 and 30, and a SIP location server 32. In the SIP application server 20, subnetworks 34 and 36 are formed between the SIP load balancer 26 and the SIP servlet engines 28 and 30. The SIP application server 20 is different from the prior art in that the subnetwork 34 can be connected not only to the SIP servlet engine 28 but also to the SIP servlet engine 30. Similarly, the subnetwork 36 can be connected not only to the SIP servlet engine 30 but also to the SIP servlet engine 28. Such interconnection between the sub-networks 34 and 36 and the SIP servlet engines 28 and 30 is based on the functions of the SIP servlet engines 28 and 30 described later.

  The SIP load balancer 26 has a function of managing, for example, the subnetwork 22 and the subnetwork 34, the subnetwork 24 and the subnetwork 36 in association with each other, and distributing the load of the SIP servlet engine. That is, the SIP load balancer 26 assigns a different logical IP address to each of the network interface cards to which the sub-networks 22 and 24 correspond, and the sub-networks 22 and 24 correspond to each of the network interface cards to which the sub-networks 34 and 36 connect. Assign the same logical IP address as the one set in step 1.

  Each of the SIP servlet engines 28 and 30 has functions such as call processing, and in particular, functions for setting interface card information based on the address port information, distributing the processing and receiving responses of processing results, and address port information Memory function is included. The components having these functions are represented by the address port management unit 38 and the memory 40, the address port management unit 42, and the memory 44 in FIG. The SIP servlet engine 28 sets an IP address assigned to the subnetwork 34 and 36 to be connected to the interface card. The SIP servlet engine 30 also sets IP addresses assigned to the sub-networks 34 and 36 to which the SIP servlet engine 30 is connected in the interface card. Thus, the SIP servlet engines 28 and 30 have the same function. The memories 40 and 44 may be stored in a database using a storage such as a hard disk. SIP servlet engines 28 and 30 connect to a SIP location server 32.

The SIP servlet engine 28 stores the address port management information shown in FIG. 2, that is, SIP transport information and SIP load balancer information in the memory 40. In this embodiment, the SIP transport information stores an LB (Load Balancer) identifier 48 as well as an IP address, a port number, a protocol type, and an activation state. Further, the SIP load balancer information stores and stores not only the load balancer address, the port number in TCP / UDP, and the activation state, but also the LB identifier 50. By providing the SIP transport information and the SIP load balancer information with the LB identifier, the address port management unit can associate the SIP transport information with the SIP load balancer information. The SIP servlet engine 28 can know the logical IP address of the response return destination SIP load balancer 26 even if a plurality of sub-networks are accommodated. The SIP application server 20 can accommodate the user information of the plurality of sub-networks 34 and 36 in the SIP servlet engines 20 and 30, respectively, and can efficiently use free resources. When a new subnetwork is added, user information of the new subnetwork can be accommodated in the existing SIP servlet engine without adding a SIP servlet engine. Thereby, existing resources can be used effectively.

  Returning to FIG. 1, the SIP location server 32 has a function of managing subscriber information and the like that are users accommodated in the sub-network 22/24. Based on this management function, the actual destination information is referenced from the SIP-URI of the destination, the corresponding destination information is obtained, and the SIP request is processed. Further, the SIP location server 32 has a storage for storing the destination SIP-URI and actual destination information, that is, user information in association with each other. The SIP location server 32 returns the obtained destination information to one of the SIP servlet engines 28 and 30 that have received the SIP request to be connected.

  Next, the operation of the SIP servlet engine 28 focusing on the SIP load balancer 26 in the multihomed network system 10 will be briefly described. This operation is not limited to the SIP servlet engine 28, and it goes without saying that the SIP servlet engine 30 can operate similarly.

  A SIP request is input to the SIP servlet engine 28 via the transport of this engine. In response to the reception of the SIP request, the process proceeds to SIP request processing (to step S10).

  The SIP servlet engine 28 processes the SIP request. The SIP request, for example, supplies the destination SIP-URI to the SIP location server 32 and refers to the destination SIP-URI to obtain user information of the destination.

  Next, the SIP location server 32 returns the obtained destination user information as a response to the SIP servlet engine 28 (step S12). In response to this response, the SIP servlet engine 28 proceeds to a process for determining the presence or absence of the SIP load balancer 26 (to step S14).

  Next, the SIP servlet engine 28 refers to the SIP transport information of the SIP request received by the address port management unit 38, and determines whether or not the SIP load balancer 26 is present (step S14). The judgment can know the load balancer address based on the LB identifier and the activation state as SIP load balancer information. Since the presence / absence of the LB identifier reflects the presence / absence of the SIP load balancer 26, it can be determined at least at this stage by using the presence / absence of the LB identifier. If it is determined that the SIP load balancer 26 does not exist in the SIP application server 20 using this condition (YES), the process proceeds to a SIP response return process (to step S16). If it is determined that the SIP load balancer 26 is present in the SIP application server 20 (NO), the process proceeds to the SIP load balancer information acquisition process (to step S18).

  In the SIP response return process, the obtained SIP response is returned in accordance with OS (Operating System) routing information (step S16).

  Also, in the SIP load balancer information acquisition process, the address port management unit 38 accesses the corresponding SIP transport information in the memory 40. In this access, SIP load balancer information is referred to based on the LB identifier 48 included in this information. This reference means that the LB identifier 48 and the LB identifier 50 are compared to obtain a corresponding load balancer address, that is, a logical IP address.

  Next, the address port management unit 38 determines whether or not the SIP load balancer 26 is activated based on the activation state included in the SIP load balancer information (step S20). In the activated state (YES), the process proceeds to a SIP response return process using the SIP load balancer information (to step S22). If it is not activated (NO), the process proceeds to a SIP response return process (to step S16). In the SIP response return process in the activated state, the SIP response is returned toward the logical IP address of the obtained SIP load balancer 26 (step S22). Assume that the SIP response to the SIP request is completed after these SIP responses are returned.

  Thus, in the SIP servlet engine, it becomes possible to accommodate users or subscribers of a plurality of different sub-networks on one system. Thereby, it is possible to effectively utilize the resources in the existing facilities without laying new facilities with accommodation.

1 is a block diagram showing a schematic configuration in which a server device of the present invention is applied to a SIP application server in a multihomed network system. It is a figure explaining the address port management information which the SIP servlet engine in the SIP application server of FIG. 1 stores. It is a flowchart explaining operation | movement of the SIP servlet engine of FIG.

Explanation of symbols

10 Multihomed network system
12-18 user terminal equipment
20 SIP application server
22, 24, 34, 36 Subnetwork
26 SIP load balancer
28, 30 SIP servlet engine
32 SIP location server
38, 42 Address Port Management Department
40, 44 memory

Claims (4)

In a server device in which information processing function blocks connected to an information device operating with a SIP (Session Initiation Protocol) protocol are distributed and arranged, the information processing function block includes the information processing function block. function block manages the first information on a plurality of sub-networks to be connected, with the transmission source user terminal to the calling destination information to the destination, is disposed in the device, it performs call processing for the SIP Information management means for managing a SIP request supplied corresponding to the designation of the destination information from a load distribution means for distributing the load of the information processing function block and a return of a SIP response to the load distribution means for the SIP request When,
IP (Internet Protocol) address, port number, protocol type, transport information indicating whether the transport is in use / not used , and transport balance information including a load balancer identifier included in the address port management information , and a load balancer address the port number for each protocol, the identifier that associates both said transport information and the load balancer information to the load balancer information possessed by providing the active state and the load balancer identifier representing the in-use / unused said load balancing unit as information Stored storage means,
Each of the information processing function blocks receives the SIP request, refers to the transport information that matches the transport information of the SIP request, among the transport information stored in the storage unit, and matches The presence / absence of the load balancer is determined based on the presence or absence of the load balancer identifier provided in the transport information, and the load balancer identifier of the transport information is determined according to the determination of the presence of the load balancer And the load balancer identifier of the stored load balancer information is compared, the load balancer address is obtained from the load balancer information of the load balancer identifier that matches, and the load balancing means is based on the activation status of the acquired load balancer information And whether or not the load balancing is activated. Returns the SIP response to the obtained load balancer address, manages the return of the SIP response to the load distribution means for the SIP request, and responds even if the plurality of sub-networks are accommodated by association obtaining know the IP address information of the return destination, the server apparatus characterized by effectively utilizing the existing resources.   2. The server device according to claim 1, wherein the server device is provided with load distribution means for connecting each of the sub-networks existing outside the device and inside the device, and distributing the load of the information processing function block. A server device as a feature.   3. The server apparatus according to claim 1, wherein a plurality of the information processing function blocks are provided and are interconnected with each of the sub-networks. The information processing function block managed by the information processing function block using an information processing function block that operates in accordance with the SIP (Session Initiation Protocol) protocol to which a plurality of sub-networks are connected and load distribution means for distributing the load of the information processing function block. A communication connection method for establishing a communication connection by returning a request supplied from the load distribution unit and a response to the request to a position having address information designated by the load distribution unit based on the information of 1. Is
IP (Internet Protocol) address included in the address port management information , port number, protocol type, transport information that indicates whether the transport is in use or not , and transport load information that includes a load balancer identifier , and load balancer address, An identifier for associating both the transport information and the load balancer information is set with respect to the load balancer information having the port number of each protocol, the active state indicating whether the load balancing means is in use or not, and the load balancer identifier as information. ,
The load balancer identifier provided in the matched transport information with reference to the transport information that matches the transport information of the request among the transport information stored in the storage means upon receiving the request A first step of determining one of presence / absence of the load balancing means based on the presence or absence of
If the it exists in the load balancing means, obtains the load balancer address from the load balancer of the load balancer identifier corresponding comparing the load balancer identifier of the load balancer information stored with load balancer identifier of the transport information a second step you,
A third step of determining whether the load distribution unit is in an activated state based on the activated state of the acquired load balancer information;
If the load distribution means is in an activated state, the fourth communication connection wherein the including that the step of returning the response towards the acquired load balancer address.

Images