JP5750933B2 - Communication system, switching hub, router and program - Google Patents

Description

  The present invention relates to a technique for supporting operation management of a LAN (Local Area Network).

  In recent years, corporations have established LANs (hereinafter referred to as local LANs) at each base such as a branch office, and connected these local LANs to an IP (Internet Protocol) network such as the Internet via a router. Building is generally done. Such a local LAN is often configured by cascading a plurality of switching hubs, and a communication terminal used by each user (for example, a company employee) is connected to any one of those switching hubs. Here, the router is a relay device that performs transfer control of packets (data transmission / reception unit in the third layer) according to the communication protocol of the third layer (network layer) in the OSI (Open Systems Interconnection) reference model. The switching hub is a relay device that performs transfer control of a frame (data transmission / reception unit in the second layer) according to the communication protocol of the second layer (data link layer) in the OSI reference model, and is also called an L2 switch.

  When performing operation management and maintenance of an in-company information system (or each local LAN that is a communication system of its constituent elements), the connection relationship between the router and each switching hub included in the local LAN in the subordinate ( It is important to understand the network topology below. For this reason, various techniques for enabling understanding of the network topology in the LAN have been proposed. For example, in the technology disclosed in Patent Document 1, a computer device called a management node is connected to a LAN whose network topology is to be grasped, and a predetermined frame is multicast / broadcast from the management node. Then, the reception of the frame causes the MAC (Media Access Control) address of the own apparatus to be written and returned in the payload portion of the response frame for the frame. Here, the MAC address is a hardware identifier that uniquely identifies each communication device in the second layer. In addition, in the technique disclosed in Patent Literature 1, when a response frame transmitted from another switching hub is transferred to each switching hub, the MAC address of the own device is added to the payload portion and transferred. Let In this way, the network topology in the LAN is grasped by tracing the array of MAC addresses written in the payload portion of the response frame returned from each switching hub.

JP 2007-212285 A

  In a large-scale LAN including a large number of switching hubs, the number of cascade connection stages is large, and the connection relationship of each switching hub is often complicated. For this reason, the importance of grasping the network topology increases as the LAN becomes larger. However, the technique disclosed in Patent Document 1 has a problem that it is difficult to apply to a large-scale LAN with many cascaded stages. The reason is as follows. In the technique disclosed in Patent Document 1, the data size of the payload portion increases each time a response frame is transferred by the switching hub. For this reason, as the number of stages of cascade connection increases, a response frame having a larger data size in the payload portion is transmitted in the LAN. That is, as the number of cascade connection stages of switching hubs increases, the load on the LAN increases due to communication for grasping the network topology. If an excessive load is applied to the LAN due to communication for grasping the network topology, the original data communication performed through the LAN (for example, data communication between communication terminals accommodated in the LAN) may be adversely affected. is there. This is the reason why it is difficult to apply the technique disclosed in Patent Document 1 to grasp the network topology in a large-scale LAN. In the technique disclosed in Patent Document 1, since the network topology is grasped when response frames transmitted from all the switching hubs are received, it takes a long time from multicasting a predetermined frame to grasping the network topology. And the response is bad. Furthermore, the technique disclosed in Patent Document 1 has a problem that a response frame is likely to be lost as the load on the LAN increases, and the network topology cannot be grasped when a response frame is lost. .

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique that makes it possible to grasp a network topology in a network under a router without applying an excessive load to the network.

  In order to solve the above problems, the present invention provides a communication system including a router and a plurality of switching hubs included in a network under the router, wherein (A) at least one of the plurality of switching hubs is: A process that has storage means and performs transfer control of the frame according to the destination of the received frame, and uniquely identifies the device that transferred the frame to the received frame in the second layer of the OSI reference model If the hardware identifier to be assigned is assigned as a transfer source identifier, after the transfer source identifier is written in the storage means, the transfer source identifier assigned to the frame is rewritten with the hardware identifier of the own device, First means for performing the transfer control on a frame in which the transfer source identifier has been rewritten, and the storage And a second means for transmitting a notification frame in which a list of transfer source identifiers stored in a stage is written in a payload portion to the router, and (B) the router adds the router to a part of the frames. A third means for assigning a unique hardware identifier as the transfer source identifier and multicasting to the network; and receiving the notification frame, from the payload portion of the notification frame to the switching hub of the notification frame And a fourth means for reading out a list of transfer source identifiers of the device that has transferred the frame, and generating topology data representing a connection relationship between the router and the switching hub based on the list. A communication system is provided.

  For example, if all of the switching hubs included in the network under the router have the first and second means, the storage means of each switching hub is transferred each time a frame with a transfer source identifier is transferred. The transfer source identifier is stored. That is, the adjacent relay device that has transferred the frame to each switching hub is specified. Here, since the router assigns a transfer source identifier to some of the frames to be multicast to the subordinate network and performs multicasting, each switching hub receives the frame and transfers the transfer source. The identifier is stored and flooded. Thereby, the adjacent relay device on the upstream side (side close to the router) for each switching hub is specified. The fact that the adjacent relay device on the upstream side is specified for each switching hub is the fact that the connection relationship between the router and each switching hub is specified.

  In addition, in the communication system of the present invention, each switching hub does not append the hardware identifier of its own device as a new transfer source identifier when transferring a frame to which a transfer source identifier is assigned, Since the identifier is rewritten, the data size of the payload portion of the frame does not increase every time the frame is transferred by the switching hub. For this reason, according to the communication system of the present invention, even if the network under the router becomes large, it is possible to specify the network topology without imposing an excessive load on the network. Although details will be described later, a switching hub included in the network under the router includes one that does not have the first and second means (specifically, a conventional switching hub). Even so, it is possible to specify a rough network topology without the conventional switching hub.

  In a more preferred aspect, the second means is characterized in that when the notification frame is transmitted to the router, a hardware identifier of the own device is assigned as a transfer source identifier and transmitted. According to such an aspect, in addition to the adjacent relay device on the upstream side of each switching hub, the adjacent relay device on the downstream side is specified.

  In another preferred aspect, the partial frame is a request frame newly defined for requesting transmission of the notification frame, and the second means indicates that the request frame has been received. The notification frame is transmitted as an opportunity. According to such an aspect, it becomes possible to define a new communication protocol (second layer communication protocol) for specifying the network topology separately from the existing communication protocol in the second layer. Here, various modes can be considered as to when the request frame is transmitted to the router. For example, there are a mode in which transmission is performed when a predetermined type of packet is received from a communication terminal included in the network, a mode in which multicast is periodically performed at a predetermined time interval, and the like. Although details will be described later, according to such an aspect, when a change occurs in the network topology, it is expected that the operation manager can be made aware of the changed network topology without delay.

  In another preferred aspect, the first means writes the port identifier of the communication port that has received the frame to which the transfer source identifier is assigned and the transfer source identifier in association with each other and writes them in the storage means. And According to such an aspect, the connection relation of the ports to the adjacent relay device (that is, which of the plurality of communication ports provided in each switching hub is connected to the adjacent relay device on the upstream side or the downstream side). In addition, the network topology in the network under the router can be specified.

  In order to solve the above-described problem, the present invention provides a storage unit and a unit that performs transfer control of the frame according to a destination of the received frame, and transfers the frame to the own device to the received frame. When the hardware identifier uniquely indicating the device that performed the transfer in the second layer of the OSI reference model is assigned as the transfer source identifier, the transfer source identifier is written in the storage unit, and is then added to the frame. Frame transfer control means for rewriting the transfer source identifier to the hardware identifier of the device itself and rewriting the transfer source identifier for the frame, and the transfer source identifier stored in the storage means That sends a notification frame in which the list is written in the payload part to a router that connects the network that includes the device and another communication network To provide a switching hub, characterized by having a frame transmitting means.

  This is because the communication system of the present invention can be constructed by combining such a switching hub and the router having the third and fourth means described above. As another aspect of the present invention, an aspect of providing a program characterized by causing a computer to function as the frame transfer control means and the notification frame transmission means is also conceivable. This is because by operating a general computer in accordance with such a program, the computer can function as the switching hub of the present invention. In addition, as a specific mode of providing such a program, a mode in which the program is written and distributed on a computer-readable recording medium such as a CD-ROM (Compact Disk-Read Only Memory), or via a telecommunication line such as the Internet A mode of distribution by downloading is conceivable.

  In order to solve the above-mentioned problem, the present invention assigns a hardware identifier uniquely indicating the router in the second layer of the OSI reference model as a transfer source identifier indicating the transfer source of the frame to some frames. A multicast means for multicasting to the subordinate network, and a notification frame transmitted from each of the one or more switching hubs included in the network, the frame being transmitted to the switching hub of the notification frame A device that receives a notification frame in which a list of hardware identifiers of other devices that have transferred is written in a payload portion, and that transfers the frame to the switching hub for each of the one or more switching hubs Identify from the list of transfer source identifiers, Providing a router, characterized by having a connection relationship identification means for generating a topology data representing the connection relation between the number of switching hubs and.

  This is because the communication system of the present invention can be constructed by combining such a router with the above-described switching hub having the first and second means. Further, as another aspect of the present invention, an aspect of providing a program that causes a computer to function as the multicast means and the connection relation specifying means is also conceivable. Specific examples of such program provision aspects are described above. That's right.

It is a figure which shows the structural example of LAN1A of 1st Embodiment of this invention. It is a figure which shows the structural example of the switching hub 50 contained in the LAN1A. 4 is a flowchart showing a flow of frame transfer control processing executed by a switching engine unit 520 of the switching hub 50. It is a figure which shows the data structure of the flame | frame transmitted / received in the same LAN1A. It is a figure which shows the structural example of the router 60 which connects the same LAN1A to another network. 3 is a diagram illustrating an example of a connection relationship among communication ports of a communication terminal 40, a switching hub 50, and a router 60. FIG. 5 is a diagram for explaining operations of a switching hub 50 and a router 60. FIG. 5 is a diagram for explaining operations of a switching hub 50 and a router 60. It is a figure which shows the structural example of LAN1B of 2nd Embodiment of this invention. It is a figure for demonstrating the network topology grasped | ascertained from the topology data produced | generated by the router 60 in the same LAN1B.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
<A: First Embodiment>
<A-1: Configuration>
FIG. 1 is a diagram illustrating a configuration example of a LAN 1A that is a communication system according to a first embodiment of this invention. The LAN 1A is a local LAN laid at a company branch, for example, and is connected to an IP network (not shown) such as the Internet by a router 60. The LAN 1A serves as a communication system in the branch, and constitutes an in-company information system in the company together with the LAN in the base laid in another branch. As shown in FIG. 1, the LAN 1A includes four switching hubs (switching hubs 50A, 50B, 50C, and 50D), two communication terminals (communication terminals 40A and 40B), and one operation terminal 30. , Including. As shown in FIG. 1, a switching hub 50A and an operation terminal 30 are connected to the router 60, and a switching hub 50B is connected to the switching hub 50A. Switching hubs 50C and 50D are connected to switching hub 50B. Communication terminal 40A is connected to switching hub 50C, and communication terminal 40B is connected to switching hub 50D.

  Each of the operation terminal 30 and the communication terminals 40A and 40B is, for example, a personal computer. The operation terminal 30 is used to cause the operation manager of the LAN 1A to perform various operations for operation management of the LAN 1A. The communication terminals 40A and 40B are accommodated in the LAN 1A by being connected to any switching hub included in the LAN 1A. Each of the communication terminals 40A and 40B accommodated in the LAN 1A is connected to another communication device (for example, another communication terminal accommodated in the LAN 1A, a communication terminal accommodated in another local LAN, or an IP network). Packet communication in accordance with the IP. Hereinafter, when it is not necessary to distinguish each of the communication terminals 40A and 40B, they are referred to as “communication terminal 40”. Although FIG. 1 illustrates the case where two communication terminals 40 are accommodated in the LAN 1A, three or more communication terminals 40 may be accommodated in the LAN 1A, and only one communication terminal 40 is included. Of course, it may be an accommodated form.

  Each of the switching hubs 50A, 50B, 50C, and 50D in FIG. 1 has the same configuration. Below, when it is not necessary to distinguish each of these four switching hubs, it describes with the "switching hub 50". The switching hub 50 is a relay device that relays data communication in a second layer (data link layer) that is a protocol layer lower than IP. The router 60 is a relay device that relays data communication in the network layer, and serves as a default gateway of the LAN 1A. That is, if the destination IP address of the packet received from the IP network is one of the communication terminals 40A or 40B, the router 60 transfers the packet to the LAN 1A, and the destination IP address is the communication terminal 40A. If it is not 40B, it is transferred to another router according to the stored contents of the routing table. Since the router 60 functions as a default gateway of the LAN 1A as described above, the LAN 1A may be hereinafter referred to as a “network under the router 60”.

  In the LAN 1A shown in FIG. 1, each relay device (that is, the router 60 and the switching hub 50) included in the LAN 1A is executed in the process of executing data communication (that is, frame transmission / reception and transfer control) according to the communication protocol of the second layer. ) To detect the adjacent relay device. Here, the adjacent relay device is another relay device that has transferred a frame to the relay device, and in the LAN 1A, is another relay device that is directly connected to the relay device. For example, in the LAN 1A, the adjacent relay device for the router 60 is the switching hub 50A, and the adjacent relay devices for the switching hub 50A are the router 60 and the switching hub 50B. The adjacent relay devices for the switching hub 50B are the switching hubs 50A, 50C and 50D, and the adjacent relay device for the switching hub 50C (or 50D) is the switching hub 50B.

  In the LAN 1A shown in FIG. 1, each switching hub 50 is caused to execute a process of notifying the router 60 of the detection result of the adjacent relay device, while the router 60 automatically performs the process based on the notification result from each switching hub 50. A process of generating topology data representing the connection relationship between the device and each switching hub 50 (that is, the network topology of the LAN 1A) is executed. The operation manager of the LAN 1A can grasp the network topology of the LAN 1A by referring to the topology data using the operation terminal 30. Hereinafter, the switching hub 50 and the router 60 that clearly show the features of the present embodiment will be mainly described.

<A-1-1: Configuration of the switching hub 50>
FIG. 2 is a block diagram showing the configuration of the switching hub 50. As shown in FIG. 2, the switching hub 50 includes a communication interface (hereinafter referred to as I / F) unit 510, a switching engine unit 520, and a storage unit 530.

  Communication I / F unit 510 has a plurality of communication ports. Each of the plurality of communication ports is connected to another communication device (in this embodiment, either the router 60, another switching hub 50, or the communication terminal 40) via a communication cable such as 100BASE-T. Is done. Each of the plurality of communication ports is assigned a unique port identifier (for example, a port number) for each communication port, and each communication port can be uniquely identified using the port identifier. The communication I / F unit 510 provides a frame received via each communication port to the switching engine unit 520, and transmits a frame provided from the switching engine unit 520 from a communication port instructed by the switching engine unit 520.

  The storage unit 530 includes, for example, a volatile memory such as a RAM (Random Access Memory) and a nonvolatile memory such as an EPROM (Erasable Programmable Read Only Memory) (none of which is shown in FIG. 2). In this nonvolatile memory, firmware (program) for causing the switching engine unit 520 to execute processing that clearly shows the characteristics of the switching hub of the present invention is stored in advance. On the other hand, the volatile memory is used by the switching engine unit 520 as a work area when executing the firmware. The volatile memory serves as a buffer for temporarily storing frames received by the communication I / F unit 510, and the volatile memory further includes a MAC address table in FIG. 2 and the adjacent relay apparatus table of FIG. 2 are stored.

  The MAC address table in FIG. 2 is not particularly different from that of a conventional switching hub. This MAC address table stores the port identifier of the communication port that received the frame in association with the transmission source MAC address of the frame received by the communication I / F unit 510. The writing of the MAC address and the port identifier to the MAC address table is performed along with the frame transfer control. For example, in the transfer process of a frame (a frame in which an ARP (Address Resolution Protocol) packet is written in the payload portion) transmitted from the communication terminal 40 for resolving the MAC address of the router 60, the transmission source MAC address (that is, the frame) , The MAC address of the communication terminal 40) and the port identifier of the receiving port of the frame are associated and written into the MAC address table. On the other hand, in the transfer process of the response frame to the frame, a process of writing the transmission source MAC address of the frame (that is, the MAC address of the router 60) and the port identifier of the reception port of the frame in association with each other is performed. Is called. The stored contents of the MAC address table are used when frame transfer control based on the destination MAC address is performed.

  In the adjacent relay device table, a connection destination identifier is stored in association with each port identifier of a plurality of communication ports included in the communication I / F unit 510. Here, the connection destination identifier indicates whether or not the connection destination of the communication port associated with the connection destination identifier is an adjacent relay device, and when the connection destination is an adjacent relay device, This is information that uniquely indicates an adjacent relay device. As described above, the adjacent relay device is another relay device (in this embodiment, the router 60 or another switching hub 50) that has transferred a frame to the switching hub 50. In this embodiment, the port identifier of the communication port connected to the adjacent relay device is associated with the MAC address of the adjacent relay device as a connection destination identifier, and other communication devices other than the adjacent relay device (in this embodiment) The communication port to which the communication terminal 40) is connected or the communication port not connected to another communication device is associated with NULL (0x00) as the connection destination identifier.

  This adjacent relay device table is generated in a state where each connection destination identifier is set to NULL when a power source (not shown) of the switching hub 50 is turned on. Then, along with relaying a predetermined specific frame, the stored contents of the table are updated (that is, the connection destination identifier corresponding to the port identifier of the communication port to which the adjacent relay device is connected is set to the adjacent Rewriting to the MAC address of the relay device). Although details will be described later, in the present embodiment, as the specific frame, a network topology specifying frame (a notification frame transmitted from each switching hub 50 to notify the router 60 of the stored contents of the adjacent relay device table) Two types of request frames that the router 60 multicasts to the subordinate LAN 1A in order to request the transmission of the notification frame are determined in advance. One of the features of this embodiment is that such an adjacent relay device table is stored in each switching hub 50, and the stored contents are updated in accordance with the transfer control of the network topology specifying frame. is there.

  The switching engine unit 520 is, for example, a CPU (Central Processing Unit). The switching engine unit 520 functions as a control center of the switching hub 50 by executing the firmware stored in the storage unit 530. The switching engine unit 520 operating according to this firmware executes the process shown in the flowchart of FIG. 3 every time a frame is received via any one of the communication ports. As shown in FIG. 3, there are two processes executed by the switching engine unit 520 in accordance with the firmware: a frame transfer control process and a notification frame transmission process. Prior to detailed description of the processing contents of these two processes, the data structure of a frame transmitted / received in the LAN 1A will be described.

  FIG. 4A is a diagram showing a data structure of a general frame transmitted / received in the LAN 1A. As shown in FIG. 4A, the frame has a header portion and a payload portion. The header part stores a transmission destination MAC address, a transmission source MAC address, and a type identifier. For example, in the case of the request frame described above, the multicast address is set as the destination MAC address, and the MAC address of the router 60 is set as the source MAC address. This is because the request frame is multicast from the router 60 toward the subordinate LAN 1A. In the case of a notification frame, the MAC address of the router 60 is set as the transmission destination MAC address, and the MAC address of the transmission switching hub 50 is set as the transmission source MAC address. This is because the notification frame is unicast from the switching hub 50 to the router 60 as described above. The type identifier is an identifier indicating the type of data (packet) stored in the payload portion of the frame. In the case of the request frame and the notification frame described above, a value (0xe812 in this embodiment) for clearly indicating that the network topology specifying data is a network topology specifying frame stored in the payload portion is set in the type identifier. The

  4B is a diagram illustrating the data structure of the payload portion of the request frame, and FIG. 4C is a diagram illustrating the data structure of the payload portion of the notification frame. As shown in FIGS. 4B and 4C, the data structure of the payload portion of the request frame and the data structure of the payload portion of the notification frame differ only in that the latter includes adjacent relay device information. As described above, the notification frame is a response that the switching hub 50 returns to the router 60 as a response to the request frame, and the stored contents of the adjacent relay device table of the switching hub 50 at the time when the notification frame is returned (that is, the The MAC address of the adjacent relay device of the switching hub 50 and the port identifier list of the communication port that received the frame transferred from the adjacent relay device) are set as the adjacent relay device information.

4B and 4C, the payload portion of the request frame and the payload portion of the notification frame include a transmission destination MAC address, a transmission source MAC address, and a frame type identifier. (Identifier indicating whether it is a request frame or a notification frame) and a transfer source identifier are written in common. As for the transmission destination MAC address and the transmission source MAC address in FIG. 4B (or FIG. 4C), the same one written in the header part of the request frame (or notification frame) is written. It is. The transmission source identifier is set with the MAC address of the transmission source at the time of transmission of the request frame (or notification frame), and the switching performed when the frame is transferred by another switching hub 50. The MAC address of the hub 50 is rewritten.
The above is the data structure of the frame transmitted / received in the LAN 1A.

  Next, processing contents of the frame transfer control process and the notification frame transmission process will be described with reference to FIG. As shown in FIG. 3, the frame transfer control process includes transfer control (step SA150) based on the destination MAC address. Here, the transfer control based on the transmission destination MAC address of the received frame is a unicast address in which the transmission destination MAC address is not a unicast address or the transmission destination MAC address is not stored in the MAC address table. In this case, the received frame is flooded, while when the destination MAC address is a unicast address stored in the MAC address table, it is stored in the MAC address table in association with the MAC address. The received frame is transmitted only from the communication port indicated by the port identifier. Here, a MAC address that is not a unicast address includes a multicast address or a broadcast address. Also, flooding means sending the received frame from all communication ports other than the communication port that received the received frame. The transfer control based on the destination MAC address is not particularly different from that in the conventional switching hub. In other words, in the conventional switching hub, only the transfer control based on the destination MAC address (the process of step SA150) is executed as the frame transfer control process.

  As shown in FIG. 3, the frame transfer control process of the present embodiment includes the processes of steps SA100 to SA120 in addition to the process of step SA150. As shown in FIG. 3, in the frame transfer control process of the present embodiment, the switching engine unit 520 first determines whether or not the received frame is a network topology specifying frame (that is, a request frame or a notification frame). The determination is made with reference to the type identifier of the header portion of the frame (step SA100). Specifically, when the value of the type identifier of the received frame is 0xe812, the switching engine unit 520 determines that the received frame is a network topology specifying frame. If the determination result in step SA100 is “No”, the switching engine unit 520 executes the process in step SA150 described above to end the frame transfer control process, and conversely, the determination result in step SA100. Is “Yes”, the processing after step SA110 is executed.

In step SA110 executed when the determination result in step SA100 is “Yes”, the switching engine unit 520 stores the network topology specifying frame in the adjacent relay apparatus table in association with the port identifier of the communication port that has received the frame. Is rewritten with the transfer source identifier written in the payload portion of the frame. Then, the switching engine unit 520 rewrites the transfer source identifier of the received network topology specifying frame with the MAC address of its own device (step SA120). As described above, when the network topology specifying frame is received, one of the features of the switching hub 50 of the present embodiment is that the processes of steps SA110 and SA120 are executed. In addition, as shown in FIG. 3, the switching engine unit 520 determines whether or not the received frame is a request frame with reference to the frame type identifier of the payload portion of the received frame (step SA130). If the result is Yes, a notification frame transmission process (step SA140) is further executed. In this notification frame transmission process, the switching engine unit 520 writes the adjacent relay device information and the transfer source identifier (MAC address of the own device) according to the stored contents of the adjacent relay device table of the own device in the payload portion. Is transmitted to the router 60 (unicast).
The above is the configuration of the switching hub 50.

<A-1-2: Configuration of the router 60>
FIG. 5 is a block diagram showing a configuration of the router 60. As illustrated in FIG. 5, the router 60 includes a communication I / F unit 610, a routing engine unit 620, and a storage unit 630. The communication I / F unit 610 has a plurality of ports like the communication I / F unit 510 of the switching hub 50, and a unique port identifier is assigned to each port. An IP network (not shown in FIG. 1), the switching hub 50A, and the operation terminal 30 are connected to three of the plurality of ports. Similar to communication I / F unit 510, communication I / F unit 610 provides a frame received via each communication port to routing engine unit 620, while a frame provided from routing engine unit 620 is provided by routing engine unit 620. Send out from the specified communication port.

  Similarly to the storage unit 530 of the switching hub 50, the storage unit 630 includes a volatile memory such as a RAM and a nonvolatile memory such as an EPROM (not shown). Firmware is stored in the nonvolatile memory. In the present embodiment, by operating the routing engine unit 620 according to this firmware, the function of the router 60 that significantly shows the features of the present embodiment is realized. On the other hand, the volatile memory is used by the routing engine unit 620 as a work area when executing the firmware, and serves as a buffer for temporarily storing received frames (or packets). . The volatile memory stores the routing table shown in FIG. 5, the MAC address table for the router 60, and the adjacent relay device table.

  The routing engine unit 620 is a CPU (Central Processing Unit) like the switching engine unit 520 of the switching hub 50. The routing engine unit 620 executes firmware stored in the storage unit 630 and functions as a control center of the router 60. The routing engine unit 620 executes packet transfer control processing, multicast processing, and connection relationship identification processing according to the firmware. The packet transfer control process is a process for transferring the packet based on the destination IP address of the packet received via the communication I / F unit 610 and the stored contents of the routing table. Since this packet transfer control process is not particularly different from that in the conventional router, a detailed description thereof will be omitted, and the following description will focus on the multicast process and the connection relationship specifying process.

  The multicast processing is processing for generating the above-described request frame, giving the MAC address of the own device as a transfer source identifier, and multicasting to the subordinate network (that is, LAN 1A). Although details will be described later, in this embodiment, this multicast processing is periodically executed every time a predetermined time T (for example, 3 seconds) elapses after the power of the router 60 (not shown) is turned on.

  In the connection relationship specifying process, first, the routing engine unit 620 is triggered by reception of a notification frame transmitted from each switching hub 50, and is associated with the port identifier of the communication port that has received the notification frame. Is rewritten with the transfer source identifier written in the payload portion of the frame. Then, the routing engine unit 620 represents the topology representing the connection relationship between the router 60 and each switching hub 50 in the LAN 1A from the stored contents of the adjacent relay device table of the own device and the adjacent relay device information notified from each switching hub 50. Generate data.

  More specifically, the routing engine unit 620 determines the own device and each switching hub 50 based on the stored contents of the adjacent relay device table of the own device and the adjacent relay device information notified from each switching hub 50. , A first pointer indicating the upstream adjacent relay device (for example, data indicating the port identifier of the communication port connected to the upstream adjacent relay device and the MAC address of the upstream adjacent relay device), and the downstream relay Bidirectional list composed of one or a plurality of second pointers indicating devices (for example, data indicating the port identifier of a communication port connected to the downstream adjacent relay device and the MAC address of the downstream adjacent relay device) Structure data of a structure is generated, and a set of these structure data is used as the topology data. Here, the upstream adjacent relay device is an adjacent relay device on the side close to the router that connects the network to which the device belongs and another network, and the downstream adjacent relay device is an adjacent relay on the side far from the router. Device. For example, in the LAN 1A shown in FIG. 1, the upstream adjacent relay device of the switching hub 50A is the router 60, and the downstream adjacent relay device is the switching hub 50B. Similarly, the upstream adjacent relay device of the switching hub 50B is the switching hub 50A, and the downstream adjacent relay devices are the switching hubs 50C and 50D. In the structure data for a device that does not have an upstream adjacent relay device such as the router 60 in FIG. 1, NULL may be set in the first pointer. Similarly, NULL may be set in the second pointer in the structure data for a device that does not have a downstream adjacent relay device, such as the switching hub 50C (or 50D) in FIG.

When a set of structure data having a bidirectional list structure as described above is used as topology data, even if two adjacent relay devices are traced using the second pointer from the upstream side, It is necessary that no contradiction occurs even if tracing is performed using a pointer. For example, when generating the structure data for each of the router 60 and the switching hub 50 shown in FIG. 1, the switching hub 50B is indicated by the second pointer of the structure data corresponding to the switching hub 50A. The switching hub 50A needs to be indicated by the first pointer of the structure data corresponding to the switching hub 50B. For this reason, for example, when the switching hub 50B is notified of the switching hub 50A as an adjacent relay device, but there is a contradiction that the switching hub 50B is not notified of the adjacent relay device as an adjacent relay device. The routing engine unit 620 generates topology data only to the extent that no contradiction occurs. This is to prevent the network topology in the LAN 1A from being misunderstood by the topology data including the contradiction.
The above is the configuration of the router 60.

<A-2: Operation>
Hereinafter, the operation of this embodiment will be described by taking as an example the case where the router 60 and each switching hub 50 have eight communication ports and are connected to each other as shown in FIG. In FIG. 6, the communication port is indicated by a black circle, and the port number of the communication port is indicated by a number with #. As shown in FIG. 6, a switching hub 50A is connected to the communication port of port number = 1 of the router 60, and the router 60 is connected to the communication port of port number = 8 of the switching hub 50A when viewed from the switching hub 50A. Has been.

  As shown in FIG. 6, a switching hub 50B is connected to the communication port of port number = 7 of the switching hub 50A. From the viewpoint of the switching hub 50B, the switching hub 50A is a communication port of port number = 5 of the switching hub 50B. It is connected to the. The switching hub 50C is connected to the communication port of port number = 4 of the switching hub 50B, and the switching hub 50D is connected to the communication port of the same port number = 6. From the viewpoint of the switching hub 50C, the switching hub 50B is connected to the communication port of port number = 3, and from the viewpoint of the switching hub 50D, the switching hub 50B is connected to the communication port of port number = 2. The communication terminal 40A is connected to the communication port of port number = 7 of the switching hub 50C, and the communication terminal 40B is connected to the communication port of port number = 8 of the switching hub 50D. Further, at the start of the operation described below, the connection destination identifier in the adjacent relay device table of each switching hub 50 is NULL cleared.

  The routing engine unit 620 of the router 60 multicasts the request frame at a time T0 when a predetermined time T has elapsed after the power (not shown) of the router 60 is turned on. The request frame multicasted from the router 60 in this manner is first received by the switching hub 50A. The switching engine unit 520 of the switching hub 50A receives the request frame via the communication port of port number = 8, and executes the processes of steps SA110, SA120, SA140, and SA150 in FIG. As described above, since the value of the type identifier in the header portion of the request frame is 0xe812, the determination result in step SA100 in FIG. 3 is “Yes”, and since the received frame is the request frame, FIG. This is because the determination result in step SA130 is also “Yes”.

  Since the process of step SA110 is executed, the connection destination identifier stored in the adjacent relay device table of the switching hub 50A in association with the port number = 8 is the transfer source identifier (that is, the router) assigned to the received frame. 60 MAC addresses). Further, the transfer source identifier of the received frame is rewritten to the MAC address of the switching hub 50A by the process of step SA120. In Step SA140, the notification frame in which the stored contents of the updated adjacent relay device table by the processing in Step SA110 is written as adjacent relay device information is unicast from the switching hub 50A to the router 60. In step SA150, the request frame in which the transfer source identifier has been rewritten by the processing in step SA120 is flooded. This is because the destination MAC address of the request frame is a multicast address as described above. The request frame flooded in this way is received by the switching hub 50B.

  Similarly, when the switching engine unit 520 of the switching hub 50B receives the request frame transferred by the upstream adjacent relay device, the processing of steps SA110, SA120, SA140, and SA150 of FIG. 3 is executed. More specifically, in the switching engine unit 520 of the switching hub 50B, the connection destination identifier corresponding to port number = 5 in the adjacent relay device table of the own device is assigned to the request frame transferred from the switching hub 50A. The notification frame is rewritten by the transfer source identifier (that is, the MAC address of the switching hub 50A), and the notification frame in which the stored contents of the rewritten adjacent relay device table is written as adjacent relay device information is unicast to the router 60. In addition, the switching engine unit 520 of the switching hub 50B performs flooding of a request frame in which the transfer source identifier is rewritten to the MAC address of the switching hub 50B. The request frames flooded in this way are received by the switching hubs 50C and 50D, respectively.

  Similarly, when the switching engine unit 520 of the switching hub 50C (or 50D) receives the request frame transferred by the upstream adjacent relay device, the processing of steps SA110, SA120, SA140, and SA150 in FIG. 3 is executed. More specifically, when the switching hub 50C receives the request frame transferred by the switching hub 50B, the switching hub 50C transfers the connection destination identifier corresponding to the port number = 3 in the adjacent relay device table of the own device from the switching hub 50B. The notification frame, which is rewritten by the transfer source identifier (that is, the MAC address of the switching hub 50B) assigned to the requested frame and the stored contents of the rewritten adjacent relay device table as the adjacent relay device information, is unidirectionally stored in the router 60. Cast.

  Similarly, the switching hub 50D also rewrites the connection destination identifier corresponding to port number = 2 in the adjacent relay device table with the MAC address of the switching hub 50B, and stores the rewritten contents of the adjacent relay device table in the adjacent relay device information. Is unicasted to the router 60. The switching hubs 50C and 50D also perform a process of flooding a request frame in which the transfer source identifier is rewritten to the MAC address of the own device. However, since the switching hub is not connected to the downstream side of the switching hubs 50C and 50D, the request frame is not transferred thereafter.

  As described above, the contents stored in the adjacent relay device tables of the switching hubs 50A, 50B, 50C, and 50D in the process of performing transfer control of the request frame multicast from the router 60 are in the state shown in FIG. A notification frame in which the contents stored in each adjacent relay device table is written as adjacent relay device information is unicast from each switching hub 50 to the router 60. In FIG. 7, the contents stored in the adjacent relay device table of each switching hub 50 are indicated by balloons (the same applies to FIG. 8).

  Next, update of the adjacent relay device table performed in accordance with notification frame transfer control will be described. When the switching engine unit 520 of the switching hub 50B receives the notification frame unicast from the switching hub 50C to the router 60 via the communication port of port number = 4, the processing of steps SA110, SA120, and SA150 in FIG. Execute. Since the value of the type identifier of the notification frame is 0xe812, the determination result in step SA100 in FIG. 3 is “Yes”, and since the reception frame is the notification frame, the determination result in step SA130 in FIG. 3 is “No”. Because it becomes.

  Since the process of step SA110 is executed, the connection destination identifier stored in the adjacent relay device table of the switching hub 50B in association with port number = 4 is the transfer source identifier (ie, switching) assigned to the received frame. The MAC address of the hub 50C is rewritten, and the transfer source identifier of the received frame is rewritten to the MAC address of the switching hub 50B by the process of step SA120. In step SA150, the notification frame in which the transfer source identifier has been rewritten by the processing in step SA120 is transmitted via the communication port with port number = 5. Similarly, when a notification frame unicast from the switching hub 50D to the router 60 is received, the switching engine unit 520 of the switching hub 50B executes the processes of steps SA110, SA120, and SA150 in FIG. As a result, the connection destination identifier corresponding to port number = 6 in the adjacent relay device table of the switching hub 50B is rewritten by the transfer source identifier (that is, the MAC address of the switching hub 50D) assigned to the received frame, and the received frame Is transmitted through the communication port with port number = 5 after rewriting the transfer source identifier.

  When the switching engine unit 520 of the switching hub 50A receives the notification frame transmitted from the switching hub 50B (or the notification frame transferred by the switching hub 50B), the switching engine unit 520 similarly performs steps SA110, SA120, and SA150 of FIG. Execute the process. The switching hub 50A rewrites the connection destination identifier corresponding to port number = 7 in the adjacent relay device table with the transfer source identifier (that is, the MAC address of the switching hub 50B) given to the notification frame received from the switching hub 50B. Then, the transfer source identifier of the notification frame is rewritten to the MAC address of the own device and transferred to the router 60. When the router 60 receives the notification frame transferred from the switching hub 50A via the communication port with the port number = 1, the router 60 is given a connection destination identifier corresponding to the port number in the adjacent relay device table of the own device. Rewrite with the existing transfer source identifier.

  As described above, in the process of performing transfer control of the notification frame unicast from the other switching hub 50 to the router 60, the contents stored in the adjacent relay device tables of the router 60 and the switching hubs 50A and 50B are as shown in FIG. The state shown in FIG. 8 is updated to the state shown in FIG. As is clear from the adjacent relay device tables shown in FIG. 8, the stored contents of these adjacent relay device tables represent the connection relationship between the router 60 and the switching hubs 50 in the LAN 1A without any contradiction.

  What should be noted here is that in the above-described operation, the adjacent relay device information written in the notification frame received by the router 60 from each switching hub 50 is stored in the content stored in the adjacent relay device table shown in FIG. Rather, it corresponds to the stored contents of the adjacent relay device table shown in FIG. Only the connection relationship between the router 60 and the switching hub 50A is represented in a consistent manner between the stored contents of the adjacent relay apparatus table of the router 60 shown in FIG. 8 and the stored contents of the adjacent relay apparatus table of each switching hub 50 shown in FIG. ing. For this reason, when the notification frame for the request frame multicasted at time T0 is received, the routing engine unit 620 generates topology data representing only the connection relationship between the router 60 and the switching hub 50A.

  When a predetermined time T further elapses from time T0, the routing engine unit 620 multicasts the request frame, and each switching hub 50 executes the same processing as described above. In each notification frame returned by each switching hub 50 as a response to the request frame transmitted from the router 60 at time T0 + T, adjacent relay device information corresponding to the stored contents of the adjacent relay device table shown in FIG. 8 is written. Yes. For this reason, when the notification frame for the request frame multicasted at time T0 + T (that is, the second multicast if the multicast at time T0 is the first) is received, the routing engine unit 620 determines that the own device and the own device Topology data representing the connection relations of all the switching hubs 50 under its control is generated. The operation manager of the LAN 1A can grasp the network topology of the LAN 1A by referring to the topology data generated in this way using the operation terminal 30.

  As described above, according to the present embodiment, the network topology in the LAN 1A is specified based on the adjacent relay device information returned from each switching hub 50. Here, the data size of the adjacent relay device information transmitted from each switching hub 50 is determined according to the number of relay devices connected to the switching hub 50, and the data according to the number of communication ports that the switching hub 50 has. No more than size. In other words, even if the number of switching hubs 50 included in the LAN 1A increases, the data size of the payload portion of the notification frame in this embodiment does not increase without limit as the number increases. For this reason, even if the scale of the LAN 1A is increased, an excessive load is not applied to the LAN 1A for specifying the network topology.

  As described above, according to the present embodiment, it is possible to allow the operation manager of the LAN 1A to grasp the network topology in the LAN 1A while avoiding an excessive load on the LAN 1A. Further, since the router 60 periodically multicasts the request frame every time the predetermined time T elapses, for example, the network topology is changed such that a switching hub of the present embodiment is newly connected to the switching hub 50D. Even so, topology data that accurately represents the network topology after the change without any contradiction is generated by multicasting the second request frame from the connection of the new switching hub. Further, in the present embodiment, since the request frame is multicast periodically, even if a frame loss occurs in the transmission process of the request frame transmitted by the N-th multicast and the response frame to the request frame, N + 1 A similar omission does not always occur in the second multicast, and not all of the specific network topologies that are periodically performed fail. In addition, according to the present embodiment, an excessive load is not applied to the LAN 1A for specifying the network topology, so that frame loss is less likely to occur in comparison with the technique disclosed in Patent Document 1. There are also features. As described above, according to this embodiment, it is possible to grasp a change in the network topology without delay while avoiding an excessive load on the LAN 1A.

<B: Second Embodiment>
FIG. 9 is a block diagram illustrating a configuration example of the LAN 1B according to the second embodiment of this invention.
As is clear from a comparison between FIG. 9 and FIG. 1, the LAN 1B is different from the LAN 1A in that it includes a switching hub 500 instead of the switching hub 50B. The switching hub 500 is a general conventional switching hub, and the switching hub 500 executes only frame transfer control based on the destination MAC address of the received frame (that is, processing in step SA150 in FIG. 3). Different from 50.

  Thus, since the switching hub 500 executes only frame transfer control based on the destination MAC address of the received frame, the transfer source identifier can be rewritten in the request frame transferred from the switching hub 50A to the switching hub 500. Instead, it is transferred to the switching hubs 50C and 50D by the switching hub 500. Therefore, the MAC address of the switching hub 50A is written in the adjacent relay device table of each of the switching hubs 50C and 50D in association with the port identifier of the communication port to which the switching hub 500 is connected.

  Similarly, when each of the switching hubs 50C and 50D unicasts notification frames to the router 60, these notification frames are transferred to the switching hub 50A by the switching hub 500 without rewriting the transfer source identifier. The switching control unit 520 of the switching hub 50A receives the notification frame transmitted from each of the switching hubs 50C and 50D via the communication port to which the switching hub 500 is connected, and is adjacent each time the notification frame is received. Update the stored contents of the relay device table. That is, the contents of the update performed according to the previously received notification frame transmitted from each of the switching hubs 50C and 50D are overwritten by the contents of the update performed according to the later received frame. .

  Therefore, for example, when the former reaches the switching hub 50A first among the notification frames returned by the switching hubs 50C and 50D in response to the request frame multicasted first from the router 60, the second time In the adjacent relay device information included in the notification frame returned by the switching hub 50A according to the request frame multicast, the port identifier of the communication port to which the switching hub 500 is connected is associated with the MAC address of the switching hub 50D. It has been. For this reason, the router 60 generates topology data representing the network topology shown in FIG. 10A when receiving a response frame for the request frame multicasted for the second time.

  On the other hand, when the latter of the notification frames returned by each of the switching hubs 50C and 50D in response to the request frame multicasted the second time from the router 60 first reaches the switching hub 50A, the third time In the adjacent relay device information included in the notification frame returned by the switching hub 50A in response to the multicast of the request frame, the port identifier of the communication port to which the switching hub 500 is connected corresponds to the MAC address of the switching hub 50C. It is attached. For this reason, the router 60 generates topology data representing the network topology shown in FIG. 10B when a response frame to the request frame multicasted for the third time is received. As described above, in the LAN 1B shown in FIG. 9, topology data representing different network topologies is generated according to the arrival order of the response frames transmitted from the switching hubs 50C and 50D. As is clear from a comparison between FIG. 10 (A) and FIG. 10 (B) and FIG. 8, a switching hub 500 appears in the network topology shown in FIG. 10 (A) and FIG. 10 (B). None of them accurately represent the actual network topology in LAN 1B.

  As described above, when the conventional switching hub is included in the network under the router 60, the connection relationship between the conventional switching hub and another device (that is, the router 60 or the switching hub 50) is accurately determined. Although it becomes impossible to grasp, it is possible to roughly grasp the connection relationship between the router 60 and the switching hub 50. In this embodiment, as shown in FIGS. 10A and 10B, one downstream communication port of the switching hub 50A (actually, the communication port to which the switching hub 500 is connected). The topology data as if the switching hub 50C and the switching hub 50D are alternately connected to each other is generated. In this way, it is impossible to actually connect the other two switching hubs alternately to one communication port of the switching hub. For this reason, the operation manager of the LAN 1B can infer that there are one or more conventional switching hubs in places where connection relationships that cannot occur in reality appear.

  Needless to say, if all of the switching hubs included in the network under the router 60 are conventional, the network topology in the network cannot be grasped, but according to the present invention, among the plurality of switching hubs constituting the LAN 1B. If at least one switching hub 50 is included, the connection relationship between the router 60 and the switching hub 50 (that is, a schematic network topology excluding a general conventional switching hub) is given to the operation manager of the LAN 1B. In addition, when an unnatural connection relationship that cannot occur in reality appears, it can be inferred that one or more conventional switching hubs exist at the corresponding location. . Also in this embodiment, the data size of the payload portion of the notification frame does not endlessly increase as the number of switching hubs 50 included in the LAN 1B increases. For this reason, according to the present embodiment, even if the LAN 1B is enlarged, it is possible for the operation manager of the LAN 1B to grasp the rough network topology of the LAN 1B without imposing an excessive load on the LAN 1B. is there.

<C: Deformation>
Although the first and second embodiments of the present invention have been described above, the following modifications may of course be added to these embodiments.
(1) In each of the embodiments described above, the router 60 periodically executes request frame multicasting at predetermined time intervals, while the switching hub 50 causes the router 60 to return a notification frame when the request frame is received. It was. However, in addition to the periodic multicast (or instead of the periodic multicast), the request frame may be multicast to the router 60 when a predetermined type of packet is received.

  For example, the request frame is transmitted to the router 60 when the ARP packet for resolving the MAC address of the own device is received. The fact that a new ARP packet has been received means that a new communication terminal has been connected to the network under the router 60, and the network topology may have changed due to the connection of the new communication terminal. If the router 60 is made to transmit a request frame in response to reception of a new ARP packet, the change can be detected quickly without waiting for the elapse of a predetermined time.

  When the router 60 also serves as a DHCP server, the request frame may be multicast when triggered by the reception of a packet requesting IP address assignment from a subordinate communication terminal. When the router 60 executes the process of periodically multicasting the request frame at a predetermined time interval and the process of multicasting the request frame when a specific packet is received, each time the latter multicast is performed. You may make it measure again progress of the said predetermined time from the time as a starting point. This is to prevent multicasting of request frames continuously at a time interval shorter than the predetermined time.

(2) In each embodiment described above, the notification frame is transmitted to the switching hub 50 when the request frame is received. However, the notification frame may be transmitted to the switching hub 50 periodically at predetermined time intervals. In this manner, in the aspect in which the notification frame is periodically transmitted to the switching hub 50, it is not necessary to cause the router 60 to periodically multicast the request frame, and the frame (multicast from the router 60 to the subordinate LAN ( A transfer source identifier may be assigned to a predetermined part of a frame (multicasted according to an existing communication protocol) for multicasting. This is because each switching hub 50 stores the MAC address of the upstream adjacent relay device in the adjacent relay device table. According to such an aspect, it becomes possible to specify a network topology using the frame transmitted / received according to the existing communication protocol.

(3) In the first embodiment described above, when the response to the request frame multicasted for the first time is received, the router 60 generates topology data representing only the connection relationship between the router 60 and the switching hub 50A immediately below it. However, topology data representing the entire network topology in the LAN 1A may be generated. By analyzing the adjacent relay device information returned from each switching hub 50 in response to the request frame multicasted for the first time, each switching hub 50 is traced from the downstream side (the adjacent relay device on the upstream side of each switching hub 50). This is because if the adjacent relay device on the upstream side of each switching hub 50 is specified, the network topology can be specified. In this manner, in the case of specifying only the adjacent relay device on the upstream side of each switching hub 50, the process of assigning the MAC address of the own device as the transfer source identifier when the notification frame is transmitted may be omitted. good.

(4) In each embodiment described above, every time a request frame is received, a notification frame in which adjacent relay device information is written in the payload portion is returned to each switching hub 50. However, if there is no change in the stored contents of the adjacent relay apparatus table between the reception of the N (natural number) request frame and the reception of the (N + 1) th request frame, the request frame received for the (N + 1) th request frame When responding, a notification frame in which data indicating that there is no change in the adjacent relay device information instead of the adjacent relay device information is written in the payload portion may be returned.

(5) In each of the embodiments described above, based on the type identifier written in the header part of the frame, it is determined whether or not the frame is for specifying the network topology. The determination may be performed depending on whether or not the In an aspect in which a transfer source identifier is assigned to a predetermined type of frames multicasted from the router 60 to the LAN 1A according to the existing communication protocol in the second layer, the type identifier of the frame includes the above A value corresponding to an existing communication protocol is set. Therefore, when the network topology is specified using a frame transmitted / received according to the existing communication protocol in the second layer, the frame is used for specifying the network topology even if the type identifier in the header part of the frame is referred to. This is because it is not possible to determine whether or not it is, but it is possible to make this determination based on the presence or absence of a transfer identifier. Similarly, in the first and second embodiments as well, it is of course possible to determine whether or not the received frame is for specifying the network topology based on the presence or absence of the transfer source identifier. When a network topology is specified using a frame transmitted / received according to an existing communication protocol, a transfer source identifier (or transfer source identifier and adjacent relay device information) is written in an empty area of the header of the frame. You can do that. Whether or not the frame is a request frame in the frame transfer control process (that is, the determination in step SA130 in FIG. 3) also depends on whether or not the adjacent relay device information is included in the payload portion of the received frame. Of course, it may be determined.

(6) In each of the above-described embodiments, the frame transfer control processing and the notification frame transmission processing that clearly show the characteristics of the switching hub of the present invention are realized by software. However, each of the frame transfer control means for executing the frame transfer control process and the notification frame transmission means for executing the notification frame transmission process is configured by hardware such as an electronic circuit, and these means are incorporated in place of the switching engine unit 520. Of course, the switching hub 50 may be configured. Similarly, for the router 60, the multicast processing and the connection relationship specifying processing may be realized by hardware such as an electronic circuit.

  1A, 1B ... LAN, 30 ... operation terminal, 40A, 40B ... communication terminal, 50A, 50B, 50C, 50D, 500 ... switching hub, 60 ... router, 510, 610 ... communication I / F unit, 520 ... switching engine unit 530, 630 ... storage unit, 620 ... routing engine unit.

Claims (8)

In a communication system including a router and a plurality of switching hubs included in a network under the router,
(A) At least one of the plurality of switching hubs is
Only for the device that has transferred the frame to the switching hub, it has a storage means for storing a hardware identifier that uniquely identifies the device that has performed the transfer in the second layer of the OSI reference model ,
A depending on the destination of the received frame process for transferring control of the frame, if the hardware identifier of the device performing the transfer of the frame has been granted as the source identifier received frame, After writing the transfer source identifier in the storage means, the transfer source identifier assigned to the frame is rewritten with the hardware identifier of the own device, and the transfer control is performed on the frame for which the transfer source identifier has been rewritten. A first means of performing;
A second means for transmitting a notification frame in which a list of transfer source identifiers stored in the storage means is written in a payload portion to the router;
(B) The router
A third means for assigning a hardware identifier uniquely indicating the router to some frames as the transfer source identifier and multicasting to the network;
Receives the notification frame, reads a list of transfer source identifiers of devices that have transferred the frame from the payload portion of the notification frame to the switching hub of the notification frame, and switches the router and the switching based on the list And a fourth means for generating topology data representing a connection relationship with the hub.   2. The communication system according to claim 1, wherein when transmitting the notification frame to the router, the second unit assigns the hardware identifier of the own device as a transfer source identifier and transmits the notification frame.   The partial frame is a request frame for requesting transmission of the notification frame, and the second means transmits the notification frame when the request frame is received. The communication system according to claim 1 or 2.   The said 3rd means multicasts the said request | requirement frame triggered by having received the packet of the predetermined kind from the communication terminal contained in the said network, or periodically. Communication system.   The communication system according to claim 4, wherein the predetermined type of packet is an ARP packet for resolving the MAC address of the router from a communication terminal included in the network.   6. The first means according to claim 1, wherein the port identifier of the communication port that has received the frame to which the transfer source identifier is assigned is associated with the transfer source identifier and written to the storage means. The communication system according to any one of the above. Storage means for storing a hardware identifier for uniquely identifying the device that has transferred the frame to the own device in the second layer of the OSI reference model only for the device that has transferred the frame ;
And means for transferring control of the frame depending on the destination of the received frame, the received frame, the hardware identifier of the device performing the transfer of the frame to the own device is applied as the transfer source identifier If the transfer source identifier is written in the storage unit, the transfer source identifier assigned to the frame is rewritten with the hardware identifier of the own device, and the transfer source identifier is rewritten. Frame transfer control means for carrying out the transfer control for,
Notification frame transmission means for transmitting a notification frame in which a list of transfer source identifiers stored in the storage means is written in a payload portion to a router connecting the network including the device and another communication network. Switching hub featuring.   Computer
  A means for performing transfer control of the frame according to the destination of the received frame, and a hardware identifier uniquely indicating the device that transferred the frame to the computer in the second layer of the OSI reference model. If it is given as a transfer source identifier, the transfer source identifier is stored in the storage means provided in the computer to store only the hardware identifier of the device that transferred the frame to the computer. A frame transfer control means for rewriting the transfer source identifier assigned to the frame after writing to the hardware identifier of the computer and performing the transfer control on the frame in which the transfer source identifier has been rewritten;
  Notification frame transmitting means for transmitting a notification frame in which the list of transfer source identifiers stored in the storage means is written in a payload portion to a router that connects the network including the computer and another network
  A program characterized by functioning as

Images