JP4421458B2 - Network node device and route information update method thereof - Google Patents

Description

  The present invention relates to a network node device that is connected to a layer 2 network and also operates as a layer 3 node, and a route information updating method for the network node device.

  In a layer 3 network, a routing protocol that operates on a multi-access network represented by OSPF (Open Shortest Path First) or RIP (Routing Information Protocol) is connected to the multi-access network by sending control packets periodically. Recognizing each other's presence between nodes, and the arrival of a predetermined time control packet, the loss of an adjacent node is recognized.

  For example, OSPF transmits a Hello packet every 10 seconds, and recognizes that the adjacent node has disappeared if the Hello packet from the adjacent node does not arrive for 40 seconds. When an adjacent node registered as the next hop (NextHop) of the route information disappears, data relay using the route is stopped, but an alternative route is calculated by the routing protocol and a new next hop is registered. Thus, the data relay can be resumed (for example, see Non-Patent Document 1).

  In RIP, an RIP response packet storing route information is transmitted every 30 seconds, and the route information is learned by RIP response packets transmitted by other node devices. For each route information, if the RIP Response packet storing the route information does not arrive from the source node device for 180 seconds, the route information is invalidated. When the RIP response packet transmitted by another node device including the route information is received after the route information is invalidated, the route information is learned and validated (for example, see Non-Patent Document 2).

  On the other hand, as a layer 2 network configuration method, Resilient Packet Ring (hereinafter referred to as RPR) is known. In this RPR, in a network in which nodes are configured in a ring shape, each node (hereinafter referred to as a ring node) advertises its physical address on the ring, and each ring node collects the advertisement information and arranges the nodes. Recognize the order (topology map). Then, when transmitting a user data packet on the ring, the topology map is searched with the physical address of the destination node, and a ring of the system close to the physical address of the destination node is selected and transmitted. In this RPR, when a failure occurs in any physical link on the ring, it is possible to perform high-speed path switching by immediately transmitting the failure information and reflecting it in ring selection (for example, non-linkage). (See Patent Document 3).

  In a configuration in which a plurality of devices (nodes) are connected to the same layer 2 network as in this RPR, when each device (node) also operates as a layer 3 node, the layer 2 network is treated as a multi-access network.

J. Moy, "Request for Comments: 2328, OSPF Version 2", The Internet Engineering Tasking Force, April 1998, [online], retrieved from the Internet: <URL: http://www.ietf.org/rfc/rfc2328 .txt> G. Malkin, "Request for Comments: 2453, RIP Version 2", The Internet Engineering Tasking Force, November 1998, [online], retrieved from the Internet: <URL: http://www.ietf.org/rfc/rfc2453 .txt? number = 2453> "IEEE Draft P802.17 / D3.3, Part 17: Resilient packet ring (RPR) access method & physical layer specifications", The Institute of Electrical and Electronics Engineers, Inc., pages 309-315, April 21, 2004

  However, if a plurality of routes passing through the ring node are prepared as a route to the outside of the ring for redundancy and the ring node is operated as a layer 3 node, if a failure occurs in the ring node on the main relay route, the RPR However, switching to an alternative route cannot be performed, and route switching by the layer 3 routing protocol is necessary. As described above, the conventional Layer 3 routing protocol technology operates on a multi-access network, and therefore, it is recognized that a loss of an adjacent node or a loss of path information is recognized by a timeout. It took time to start acquiring routes. Therefore, there is a problem that it takes a long time until the data relay is stopped and restarted.

  The present invention has been made in view of the above, and in a network node device that performs routing control using a layer 3 routing protocol that operates on a multi-access layer 2 network that manages a topology state by a physical address, An object of the present invention is to obtain a network node device capable of shortening the path switching time when a failure occurs in another node device. Another object of the present invention is to obtain a route information update method for a network node device when a failure occurs in a multi-access layer 2 network.

To achieve the above object, a network node device according to the present invention is a network node device that constitutes a network node device of a layer 2 network and performs data relay processing in accordance with a layer 3 routing protocol, and has a predetermined time interval. In addition to the above, when a change occurs in the link state, topology information including a link state with a node device adjacent to the own node device is transmitted, and based on the topology information received from another node device, the layer 2 network A logical address used in layer 3 corresponding to the physical address is obtained from a layer 2 control means for generating topology map information indicating a connection state and a physical address used in layer 2 for a node device in the layer 2 network. The physical address and logical address And ARP table generating means for generating an ARP table for associating the scan, and generates a routing table based on Layer 3 routing protocols, and layer 3 routing protocol processing means for performing a relay processing of data in accordance with the routing table, the Layer 3 network Neighbor registration table storage means for storing a neighbor registration table including a node device adjacent to the above-mentioned own node device, and a change in the topology map information is detected, and a physical address of the node device lost from the topology map information is and topology change detecting means for notifying the topology information change detection notification to the connection information update means including, acquires the logical address corresponding to the physical address of the topology information change detection notification from the ARP table, the having a logical address obtained Comprising a connection information updating means for deleting the entry in the driver registration table, wherein the Layer 3 routing protocol processing unit detects an update of the neighbor registration table, characterized by regenerating the routing table.

  According to the present invention, when a failure occurs in the layer 2 network, the topology change detecting means detects a change in the topology information of the lower layer, and based on this detection, the connection information updating means detects the node connection information on the layer 3 routing protocol, for example By updating the contents of the neighbor registration table, there is an effect that the route information in the layer 3 network can be quickly updated in accordance with the change in the topology information of the layer 2 network.

  Exemplary embodiments of a network node device and a route information updating method thereof according to the present invention will be explained below in detail with reference to the accompanying drawings.

Embodiment 1 FIG.
FIG. 1 is a diagram schematically showing an example of a network configuration using a ring-shaped layer 2 network in which the network node device according to the present invention is arranged, and FIG. 2 shows the implementation of the network node device according to the present invention. 3 is a block diagram showing a functional configuration of the first embodiment, and FIG. 3 is a logical configuration diagram of the network when the network configuration of FIG. 1 is viewed at the layer 3 level.

  In the first embodiment, the layer 2 (second layer) network 110 has a bidirectional double ring configuration. That is, each of the network node devices (hereinafter also referred to as node devices) 100 to 104 has a configuration in which the communication lines 111 and 112 are connected to each other in a ring shape. Of the two communication lines, one communication line 111 transmits data clockwise (clockwise) and is referred to as a 0 system or 0 system ring, and the other communication line 112 is counterclockwise ( It is assumed that the data is transmitted counterclockwise (1) or 1-system ring. The node devices 100 to 104 constituting the layer 2 network have the same layer 2 control method and layer 3 (third layer) control method, respectively, and have a function of transmitting and receiving data.

  In the network configuration shown in FIG. 1, a layer 2 network 110 having a bidirectional double ring configuration formed by node devices 100 to 104 having a layer 2 control method and a layer 3 control method, and the node device of this layer 2 network 110 An information processing terminal device (hereinafter simply referred to as a terminal) A connected to 100, and a node device 102, 103 and a layer 3 node device 105 having only the same layer 3 control method as the node devices 100 to 104, And a terminal B connected to the layer 3 node device 105. In the first embodiment, RPR is used as the layer 2 control method, and path control is performed based on the connection information between the network node devices 100 to 104 having the layer 3 node device 105 or the layer 3 control method and the network. It is assumed that OSPF is used as the method to be performed.

  The network configuration shown in FIG. 1 is as shown in FIG. 3 when observed at the layer 3 level. That is, the layer 2 ring network having a bidirectional double ring configuration is a network (multi-access network) N4 that connects node devices to each other. Further, the terminal A is accommodated in the network N5 set in the node device 100 of the network N4, and the terminal B is accommodated in the network N1 set in the layer 3 node device 105. Further, the layer 3 node device 105 and the node device 102 of the network N4 and the layer 3 node device 105 and the node device 103 of the network N4 are connected by networks N2 and N3, respectively.

  In the following, a specific example will be described with reference to FIGS. 1 and 3 as appropriate. FIG. 4 shows addresses, identification information, and the like set in each node necessary at that time. 4 shows the physical addresses of the node devices (network node devices 100 to 104 and the layer 3 node device 105), the IP (Internet Protocol) addresses of the node devices (network node devices 100 to 104), and the node devices ( A router ID (Identification) for identifying the network node devices 100 to 104 and the layer 3 node device 105) is shown.

  The network node devices 100 to 104 arranged in the layer 2 network 110 include the first link control unit 11 and the second link control unit 12 that control the duplex communication lines 111 and 112, and the data processing in the layer 2 A layer 2 control unit 13 that performs control, a topology map storage unit 14 that stores a topology map of the layer 2 network, an ARP table generation unit 15 that generates an ARP (Address Resolution Protocol) table of the layer 2 network, and an ARP table ARP table storage unit 16 for storing, layer 3 routing protocol processing unit 17 for performing processing necessary for packet relay according to the layer 3 routing protocol, and neighbor registration for storing information on adjacent network node devices at the layer 3 level Table storage 18 and A routing table storage unit 19 that stores a routing table, a topology change detection unit 20 that detects a change in topology information in the layer 2 network 110, and connection information that updates connection information between the node devices 100 to 104 based on the topology information. And an updating unit 21.

  The first link control unit 11 has a function of controlling the first link, which is an interface serving as an input from the 0 system ring and an output to the 1 system ring. Further, the second link control unit 12 has a function of controlling the second link which is an interface serving as an input from the 1-system ring and an output to the 0-system ring. For example, when a failure occurs in a node device connected adjacent to the second link side or a communication line with this node device, the first link control unit 11 determines that the link failure has occurred in the layer 2 control unit 13. Notify

  The layer 2 control unit 13 has a function of exchanging topology information with other node devices configuring the layer 2 network 110, generating a topology map, and managing the generated topology map. Specifically, the layer 2 control unit 13 determines whether the physical address information, which is the physical address used in layer 2 of the own node device, and whether the connection between the own ring node device and the adjacent node device is normal. Based on the topology information 113, 114 periodically transmitted to the 0-system ring 111 and the 1-system ring 112, respectively, and topology information 113, 114 including link state information indicating the link state, respectively. A topology map is generated by grasping the arrangement order of the node devices on the rings 111 and 112. Further, the layer 2 control unit 13 monitors the link state with the adjacent node device, and when a failure occurs in any of the physical links on the rings 111 and 112 and the link state changes. Also, it has a function of immediately transmitting the topology information 113 and 114 reflecting the link state.

  The topology map storage unit 14 stores the topology map generated by the layer 2 control unit 13. FIG. 5 is a diagram illustrating an example of the configuration of the topology map. The topology map 14a shown in FIG. 5 is generated by the node device 100, and shows a case where the link state of the entire layer 2 network 110 is normal. The topology map 14a includes items of “physical address”, “0 system TTL (Time To Live)”, “1 system TTL”, “first link state”, and “second link state”. The “physical address” indicates a physical address of the node device configuring the ring-shaped layer 2 network 110. “0-system TTL” indicates the number of hops to the destination via the 0-system ring, that is, the clockwise ring (communication line) 111, and “1-system TTL” indicates the 1-system ring, that is, the counterclockwise ring ( The number of hops to the destination in the case of passing through the (communication line) 112 is shown. The “first link state” represents the link state of the first link (that is, the input side of the 0-system ring and the output side of the 1-system ring), and the “second link state” represents the second link (that is, 1 The link state on the input side of the system ring and the output side of the 0-system ring) is shown.

  For example, the topology information about the node device 101 is stored in the entry 502 of the topology map 14a. The number of hops from the node device 100 is “1” in the case of the 0 system ring 111, and the 1 system ring In the case of 112, “4” is indicated. Moreover, it is shown that the link states of the first link and the second link are both normal. The topology information stored in the other entries 501, 503 to 505 can be similarly interpreted. An entry in which the values of the 0-system TTL and the 1-system TTL are “0” is an entry indicating the own node device. From this topology map 14a, the node devices 101 to 104 of “MAC1”, “MAC2”, “MAC3”, “MAC4” are arranged in a clockwise direction from the node device 100 of the physical address “MAC0”. It is understood that the node device 100 of “MAC0” is connected after the node device 104 to form a ring.

  The ARP table generation unit 15 performs processing for acquiring the physical address of the node device corresponding to the IP address from the network to which the node device belongs (in the layer 2 network 110), and the result is the ARP in the ARP table storage unit 16. Has the function of storing in a table. For example, when it is desired to transmit data to the node device corresponding to the acquired IP address, the ARP request packet in which the IP address is set when the ARP table is searched based on the IP address but the corresponding physical address is not found. Are transmitted in the layer 2 network 110. Then, the ARP response packet from the target node device is received, and the set of the IP address and the physical address is stored in the ARP table.

  The ARP table storage unit 16 stores an ARP table indicating an association between an IP address (logical address) used in the layer 3 network and a physical address used in the layer 2 network in the interface of each of the node devices 100 to 104. To do. This ARP table shows correspondence between IP addresses and physical addresses of devices in the same subnet as the own node device. FIG. 6 is a diagram illustrating an example of the ARP table. In FIG. 6, the ARP table 16a generated by the node device 100 is shown. The ARP table 16a includes an item “IP address” that stores the IP addresses of the node devices 100 to 104 in the layer 2 network 110 that are in the same subnet, and an item “physical address” that stores a physical address corresponding to the IP address. It is comprised including. The ARP table 16a is referred to when relaying a packet, and is used to acquire a physical address used when transmitting the packet to the destination device or the next hop.

  The layer 3 routing protocol processing unit 17 manages information of adjacent node devices at the layer 3 level to generate a neighbor registration table, and performs data based on network connection information transmitted by the adjacent node device at the layer 3 level. It has a function of generating a routing table for determining a (packet) transfer route. The layer 3 routing protocol processing unit 17 has a function of regenerating the routing table based on the network connection information when the contents of the neighbor registration table are updated by the connection information updating unit 21 described later. Furthermore, it has a function to relay the received packet from its destination IP address based on the routing table.

  Here, a neighbor registration table generation process by the layer 3 routing protocol processing unit 17 and a scrutiny process for detecting an update of the contents of the neighbor registration table will be described. In the OSPF used in the first embodiment, an OSPF Hello packet is used to generate a neighbor registration table, and the layer 3 routing protocol processing unit 17 periodically transmits an OSPF Hello packet. FIG. 7 is a diagram schematically showing the contents of an OSPF Hello packet. The OSPF Hello packet 200 includes an OSPF data part 205 and an IP header part 201. The OSPF data unit 205 includes a Hello information unit 210 constituting the Hello packet 200 and an OSPF header unit 206 added to the Hello information unit 210. The OSPF header section 206 includes an area 207 including OSPF version information, a router ID (denoted as RouterID in the figure) 208 indicating a node device (router) that executes the layer 3 routing protocol, an OSPF packet checksum, and the like. And a region 209 including The IP header section 201 includes an area 202 including IP version information, a source IP address 203, and a destination IP address 204.

  FIG. 8A is a flowchart of a procedure for generating a neighbor registration table by the layer 3 routing protocol processing unit. The layer 3 routing protocol processing unit 17 receives the OSPF Hello packet 200 periodically transmitted from the adjacent node device at the layer 3 level (step S11), and recognizes the presence of the adjacent node device. Since the received OSPF Hello packet 200 stores the source IP address 203 of the node device that transmitted this packet and its router ID 208, the result is stored in the neighbor registration table in the neighbor registration table storage unit 18. (Step S12). Then, the last reception time of the OSPF Hello packet 200 is updated (step S13), and the neighbor registration table generation process ends.

  FIG. 8-2 is a flowchart illustrating a procedure of a neighbor registration table scrutiny process by the layer 3 routing protocol processing unit. The layer 3 routing protocol processing unit 17 loops in order to detect the change from the first entry in the neighbor registration table (step S21). For the first entry, it is determined whether or not a predetermined time set by RouterDeadInterval has passed since the last reception time of the OSPF Hello packet 200 (step S22). This RouterDeadInterval is 3 to 4 times the HelloInterval that is the transmission interval of the OSFP Hello packet 200 in order to operate the OSPF stably even when the network is temporarily congested and the OSPF Hello packet 200 is discarded. Is set to the value of If a predetermined time has elapsed from the last reception time of the OSPF Hello packet 200 (Yes in step S22), the entry is deleted from the neighbor registration table (step S23). Thereafter, or when a predetermined time has not elapsed since the last reception time of the OSPF Hello packet 200 (No in step S22), it is determined whether or not the examined entry is the last entry in the neighbor registration table (step S22). S24). If the entry is not the last entry (No in step S24), the process proceeds to the next entry in the neighbor registration table (step S25), and the processing from step S22 is repeatedly executed for this entry. If the examined entry is the last entry (Yes in step S24), the neighbor registration table scrutiny process ends.

  Further, the routing table generation processing by the layer 3 routing protocol processing unit 17 is the same as the routing table generation processing in the case of using the conventionally known OSPF, and the network connection information is exchanged between the node devices. Based on the information, a route to the entire network is generated and stored in the routing table in the routing table storage unit 19. However, the first embodiment is characterized in that the layer 3 routing protocol processing unit 17 performs a routing table generation process even when a change in the contents of the topology map 14a is detected.

  The neighbor registration table storage unit 18 has a function of storing the neighbor registration table generated by the layer 3 routing protocol processing unit 17. FIG. 9 is a diagram illustrating an example of the contents of the neighbor registration table stored in the neighbor registration table storage unit. The neighbor registration table 18a of FIG. 9 shows the one generated by the node device 100. As described above, the neighbor registration table 18a includes an item of “IP address” indicating the transmission source IP address of the adjacent node device and “router ID” indicating identification information given to the adjacent node device. Is done. For example, the entry 601 indicates that the node device 101 whose IP address is “10.0.0.101” and whose router ID is “127.0.0.101” is an adjacent node device at the layer 3 level. Show. The same applies to the other entries 602-604.

  The routing table storage unit 19 has a function of storing the routing table generated by the layer 3 routing protocol processing unit 17. FIG. 10 is a diagram illustrating an example of a data configuration of the routing table stored in the routing table storage unit. In FIG. 10, a routing table 19a generated by the node device 100 is shown. The routing table 19a routes to a network designated by a combination of a “destination address” indicating a destination network of data to be relayed, a “destination mask” indicating its subnet mask, and a “destination address” and “destination mask”. In order to do so, it includes an item of “next hop” indicating the IP address of the node device to be transmitted next and “type” which is the type of the layer 3 routing protocol used. For example, in the entry 701 of FIG. 10, the next hop of the route to the network N1 (indicated as “NetworkN1” and “MaskN1” in the figure) learned by the node device 100 by OSPF is “10.0.0. 103 ”, that is, the IP address of the node device 103. The same applies to the other entries 702 to 705.

  The topology change detection unit 20 detects the disappearance of the node device existing so far from the change of the topology map 14a stored in the topology map storage unit 14, and the topology information indicating the change of the topology information to the connection information update unit 21 It has a function of notifying a change detection notification. The topology information change detection notification includes the physical address information of the lost node device.

  When the connection information update unit 21 receives the topology information change detection notification from the topology change detection unit 20, the connection information update unit 21 acquires an IP address corresponding to the physical address information included in the notification from the ARP table 16a, and a node corresponding to the IP address. It has a function of deleting device information from the neighbor registration table 18a.

  As described above, in the first embodiment, the topology change detection unit 20 and the connection information update unit 21 are provided so that the change in the topology information generated in the layer 2 network 110 can be quickly used for packet transfer in the layer 3 network. This is reflected in the neighbor registration table 18a. In response to the update of the neighbor registration table 18a, the layer 3 routing protocol processing unit 17 performs a process of regenerating the routing table 19a, so that the relay destination is quickly changed at the layer 3 level.

  Here, the relay processing of packets of the node devices 100 to 104 in the layer 2 network 110 having the above-described configuration, and the route information update processing in the node devices 100 to 104 when an abnormality occurs in the node device of the layer 2 network 110 Will be described.

  FIG. 11 is a flowchart illustrating a procedure of packet relay operation processing by the node device when the layer 2 network is normal. First, when receiving the IP packet (step S31), the layer 3 routing protocol processing unit 17 of the node devices 100 to 104 in the layer 2 network 110 searches the routing table 19a using the destination IP address of the IP packet. That is, the “destination address” and “destination mask” corresponding to the destination IP address of the IP packet are searched using the routing table 19a (step S32), and the IP address of the next hop node device is acquired (step S33). ).

  Next, the layer 3 routing protocol processing unit 17 searches the ARP table 16a for a physical address corresponding to the acquired IP address of the next hop node device (step S34). If there is no entry in the ARP table 16a corresponding to the acquired IP address (No in step S34), the ARP table generating unit 15 sends an ARP request packet targeting the IP address of the next hop node device to the network. N4 (layer 2 network 110) is transmitted (step S35). An ARP response packet as a response is transmitted by the node device having received the ARP request and having a corresponding physical address. When receiving the ARP response packet (step S36), the ARP table generation unit 15 of the node devices 100 to 104 acquires the physical address of the transmission source device included in the ARP response packet (step S37), and uses the transmission source protocol address. It is stored in the ARP table 16a in association with the IP address of a node device of a certain next hop (step S38).

  After that, or when an entry corresponding to the IP address acquired in step S34 exists in the ARP table 16a (in the case of Yes in step S34), the layer 3 routing protocol processing unit 17 acquires it from the ARP table 16a in step S33. A physical address corresponding to the IP address of the node device obtained is acquired (step S39), and a process of transmitting a relay IP packet is performed. Next, the layer 2 control unit 13 stores the received relay IP packet in the Ethernet (registered trademark) frame (step S40), and obtains its destination physical address in the physical address of the node device serving as the next hop acquired in step S39. The ring is set (step S41), the ring for transmitting the frame is determined with reference to the topology map 14a, and transmitted within the network N4 (step S42). Thus, the packet relay operation processing by the node devices 100 to 104 is completed.

  In this way, when the layer 2 network 110 is normal, in order to relay the packet to the destination specified by the layer 3 IP address, the correspondence relationship with the physical address of the layer 2 is resolved by the ARP table 16a. Data (packet) is transmitted on the physical network (layer 2 network 110) by designating the physical address.

  Next, route information update processing by the node device when an abnormality occurs in the layer 2 network 110 will be described. FIG. 12 is a flowchart illustrating a procedure of operation processing of a node device adjacent at a layer 2 level to a node device in which a failure has occurred in the layer 2 network. The layer 2 control unit 13 of the node device adjacent to the failed node device at the layer 2 level detects the link break of the link connected to the failed node device (step S61). A topology information packet indicating that the link is broken is transmitted from the output side ring of the reverse link (step S62). For example, when a failure occurs in the node device on the second link side, the layer 2 control unit 13 sends a topology information packet including link state information indicating that the second link is broken on the first link side. Transmit to system 1 ring 112. Similarly, when a failure occurs in the node device on the first link side, the layer 2 control unit 13 sends a topology information packet including link state information indicating that the first link is broken to the second link side. Is transmitted to the 0-system ring 111. These topology information packets are relayed along the rings 111 and 112 and reach all the node devices 100 to 104 on the layer 2 network 110 constituting the ring network. Thereby, the processing of the node device adjacent to the node device in which the failure has occurred is completed.

  FIG. 13 is a flowchart showing the operation procedure of the path information update process of the node device in the layer 2 network when a failure occurs. First, when receiving the topology information packet transmitted from the node device adjacent to the failed node device (step S81), the layer 2 control unit 13 of the node devices 100 to 104 updates the topology map, As a result, the failed node device is removed from the topology map 14a (step S82). Next, the topology change detection unit 20 detects the node device lost due to the update of the above-described topology map 14a (step S83), and sends a topology information change detection notification including the physical address information of the lost node device to the connection information update unit 21. (Step S84).

  Upon receiving the topology information change detection notification, the connection information update unit 21 searches the ARP table 16a for the IP address of the node device corresponding to the physical address information included in the notification, and the node information corresponding to the physical address information. An IP address is acquired (step S85). Next, the connection information updating unit 21 searches the acquired IP address from the neighbor registration table 18a, and deletes the entry corresponding to the node device having the IP address from the neighbor registration table 18a (step S86). Thereafter, the layer 3 routing protocol processing unit 17 detects that the node device information has been deleted from the neighbor registration table 18a (step S87), and recalculates the route to generate the routing table 19a (step S88). The result is stored in the routing table storage unit 19, and the route information update processing of the node device when a failure occurs is completed.

  Next, regarding the specific operation processing of the node device at the time of normality and failure occurrence described above, the node device of the packet transmitted from the terminal A to the terminal B in the network configuration shown in FIG. 1 (FIG. 3) The relay process in 100 will be described as an example.

  First, a description will be given of a packet relay process from the terminal A to the terminal B of the node device 100 in a normal case where no failure has occurred in the node devices 100 to 104 of the layer 2 network 110 (network N4). First, when receiving an IP packet from the terminal A, the layer 3 routing protocol processing unit 17 of the node device 100 creates an entry having a destination address “NetworkN1” and a destination mask “MaskN1” corresponding to the destination IP address of the packet in FIG. From the routing table 19a. As a result, “10.0.0.103” is acquired from the entry 701 as the IP address of the next-hop node device. This IP address “10.0.0.103” indicates the node device 103.

  Next, the layer 3 routing protocol processing unit 17 retrieves the physical address corresponding to the acquired IP address “10.0.0.103” of the node device 103 from the ARP table 16a of FIG. Is obtained. Thereafter, the layer 2 control unit 13 stores the relay IP packet from the terminal A in the Ethernet (registered trademark) frame, sets the destination physical address to the physical address “MAC3” of the node device 103, and Transmit within a network N4. At the time of this frame transmission, the layer 2 control unit 13 refers to the entry 504 in the topology map 14a in FIG. 5 and selects one system with a small number of hops from the physical address “MAC3” of the destination node device 103. Ethernet (registered trademark) frames are transmitted in the order of the node device 100 → the node device 104 → the node device 103. Thus, the packet relay operation process when the node device 100 is normal ends.

  As described above, in order to relay the packet to the destination specified by the layer 3 IP address, the correspondence between the IP address and the layer 2 physical address is resolved by the ARP table 16a, and data is transmitted from the physical address. Data is transmitted on a physical network (ring network) by determining an optimal link.

  Next, a packet relay process from the terminal A to the terminal B of the node device 100 when a failure occurs in the node device 103 of the layer 2 network 110 will be described. FIG. 14 is a diagram illustrating an example of a network configuration when an abnormality occurs in the layer 2 network in FIG. When the node apparatus 103 is stopped due to a failure as shown in FIG. 14 from the state of FIG. 1, the node apparatus 103, which is the next hop when the packet is relayed from the node apparatus 100 to the network N1, is used for packet relay. Unusable. Therefore, communication from terminal A to terminal B is stopped. At this time, the layer 2 control unit 13 of the node device 102 adjacent to the node device 103 in which the failure has occurred detects the disconnection of the second link connected to the node device 103, and indicates that the second link is a link failure. A topology information packet 116 including the displayed link state information is transmitted to the 1-system ring 112. Similarly, the layer 2 control unit 13 of the node device 104 adjacent to the node device 103 in which the failure has occurred also detects the disconnection of the first link connected to the node device 103, and indicates that the first link is a link failure. A topology information packet 115 including the displayed link state information is transmitted to the 0-system ring 111. These topology information packets 115 and 116 are relayed along the rings 111 and 112 and reach all the node devices 100 to 104 on the ring network.

  Upon receiving the topology information packets 115 and 116, the layer 2 control unit 13 of the node device 100 updates the topology map according to the contents of the topology information packet, and as a result, the node device 103 that has failed from the topology map 14a. Entry 504 is removed. FIG. 15 is a diagram showing an example of a topology map generated after the occurrence of an abnormality in the layer 2 network, and shows a topology map 14b generated by the device 100 as in FIG. The entry 513 of the topology map 14b represents that the second link side of the node device 102 whose physical address is “MAC2” is broken, and the entry 514 is a node whose physical address is “MAC4”. This indicates that the first link side of the device 104 is disconnected. The item “0-system TTL” indicates that the “MAC0” node device 100 cannot be connected to the “MAC4” node device 104 in the 0-system ring 111. Similarly, the item “1 system TTL” indicates that connection from the node device 100 of “MAC0” to the node device 101 of “MAC1” and the node device 102 of “MAC2” cannot be established in the 1 system ring 112. Further, as apparent from the topology map 14a of FIG. 5, the entry “504” of “MAC3” indicating the node device 103 has been deleted, and the link state in the direction via the node device 103 has also been changed.

  Next, the topology change detection unit 20 of the node device 100 detects the lost node device 103 from the changes in the topology maps 14a and 14b shown in FIGS. 5 to 15, and the physical address information “ The topology information change detection notification including “MAC3” is notified to the connection information update unit 21. The connection information updating unit 21 retrieves the IP address of the node device 103 corresponding to the physical address information “MAC3” included in the topology information change detection notification from the ARP table 16a of FIG. 0.0.103 ". Further, the connection information updating unit 21 searches the neighbor registration table 18a of FIG. 9 from the acquired IP address of the node device 103, and deletes the IP address, that is, the entry 603 corresponding to the node device 103 from the neighbor registration table 18a. FIG. 16 is a diagram illustrating an example of a neighbor registration table from which entries related to the node device 103 are deleted.

  When the layer 3 routing protocol processing unit 17 detects that the entry 603 of the node device 103 has been deleted from the change in the neighbor registration tables 18a and 18b shown in FIGS. 9 to 16, the layer 3 routing protocol processing unit 17 recalculates the route and performs the routing table. And the result is stored in the routing table storage unit 19. FIG. 17 is a diagram illustrating an example of a recalculated routing table. In this routing table 19b, the next hop to the destination address “network N1” (the group represented by “NetworkN1” and “MaskN1” in the figure) in which the terminal B is accommodated is shown in the entry 711 as a failure. The IP address “10.0.0.103” (see entry 701 in FIG. 10) of the node device 103 before the occurrence is changed to the IP address “10.0.0.102” of the node device 102.

  After the next hop is changed, the node device 100 relays the IP packet transmitted from the terminal A to the terminal B to the node device 102 which is the new next hop according to the routing table 19b. Communication from terminal A to terminal B that was interrupted due to the occurrence of is resumed. At this time, the layer 2 control unit 13 of the node device 100 refers to the topology map 14a in FIG. 15 and uses the 0 system for data transmission to the physical address “MAC2” of the node device. Thus, transmission is performed using the 0-system ring 111 in the order of the node device 100 → the node device 101 → the node device 102.

  In the above description, the ARP table generation unit 15 transmits an ARP request to each adjacent node device on the layer 2 network 110 when the adjacent node device is recognized, and periodically transmits an ARP request. You may make it do. As a result, an entry in the ARP table 16a of the adjacent node device on the layer 2 network 110 always exists, so that it is not necessary to execute the ARP procedure when the topology information change detection notification is issued, and the corresponding adjacent The node device information (IP address) can always be acquired, and the path can be changed immediately and reliably.

  According to the first embodiment, the topology map change is immediately reflected in the neighbor registration table, so that the new route information is established and relayed after the node device constituting the layer 2 network 110 fails. The time until the restart is remarkably shortened compared to the time until the layer 3 routing protocol processing unit 17 performs the entry switching in the neighbor registration table, that is, the time for switching the route due to the passage of time set in RouterDeadInterval. Has the effect of being able to.

Embodiment 2. FIG.
In the first embodiment, the case where OSPF is used as the layer 3 routing protocol has been described. However, in this second embodiment, as the layer 3 routing protocol, RIP that performs route control by exchanging route information of layer 3 nodes is used. The case where it is used will be described.

  FIG. 18 is a block diagram showing a functional configuration of the network node device according to the second embodiment of the present invention. The network node devices 100 to 104 are node devices used in the layer 2 network 110 (network N4) in FIGS. 1 and 3 of the first embodiment, and control the duplex communication lines 111 and 112. The first link control unit 11, the second link control unit 12, the layer 2 control unit 13 that controls data processing in layer 2, the topology map storage unit 14 that stores the topology map, and the layer 2 network 110 An ARP table generation unit 15 that generates an ARP table, an ARP table storage unit 16 that stores an ARP table, a layer 3 routing protocol processing unit 17 that generates a routing table based on RIP and performs packet relay processing, and a routing A routing table storage unit 22 for storing the table; A topology change detection unit 20 for detecting a change in the topology information in the second network, and includes a routing information updating unit 23 for updating the routing information between network node apparatus, the by a change of the topology map. Note that components having the same functions as those of the first embodiment are denoted by the same names and the same reference numerals and description thereof is omitted, and only processing units different from those of the first embodiment will be described below.

  The routing table storage unit 22 has a function of storing a routing table generated by the layer 3 routing table protocol processing unit. In the second embodiment, a routing table generated based on RIP is stored. FIG. 19 is a diagram illustrating an example of a routing table. In FIG. 19, a routing table 22a generated by the node device 100 is shown. In addition to the items shown in the routing table 19a shown in FIG. 10 of the first embodiment, an item “metric” which is the number of hops to the network specified by the combination of “destination address” and “destination mask” is added. Consists of including. For example, in the entry 721, the next hop of the destination represented by the combination of the destination address “NetworkN1” and the destination mask “MaskN1” is “10.0.0.103”, and the layer 3 routing that learned the network path of the entry The protocol type is “RIP”, and the metric to the next hop is “3”. The same applies to the other entries 722 to 725.

  The layer 3 routing protocol processing unit 17 creates a routing table based on route information exchanged with adjacent network node devices at the layer 3 level based on RIP, and performs a relay process on received packets Have Specifically, the routing table creation function periodically transmits a RIP Response message storing a route entry that is the content of each entry in the routing table of its own node device (every 30 seconds), and other nodes The route entry destination information stored in the RIP response packet received from the device is compared with the routing table of the own node device, and processing for adding or updating the contents of the routing table is performed. In addition, a route invalidated by the route information update unit 23 described later is learned from the RIP response packet received from another node device as the next hop to the destination network, and the IP address of the node device that transmitted the message is learned. The route entry information included in this packet is used to update the corresponding entry in the routing table to validate the route.

  FIG. 20 is a diagram illustrating an example of the structure of a RIP Response packet used in RIP. The RIP response packet 300 includes a header part 301 and route entry parts 310 and 320. The header section 301 includes RIP headers such as an IP header 302, a UDP header 303, an area 304 for storing a command field, and an area 305 for storing a field such as Version. The route entry units 310 and 320 include address family fields 311 and 321, IP address fields 312 and 322 indicating the IP address of the destination network, subnet mask fields 313 and 323 indicating the subnet mask of the destination network, and the next to the destination network. It includes a next hop (NestHop) field 314, 324 indicating a hop, and a metric field 315, 325 indicating a metric to the destination network. The route entry units 310 and 320 store the contents of routing table entries, and store a maximum of 25 route entries. Here, a case where two route entries 310 and 320 are included is illustrated.

  FIG. 21 is a flowchart showing the procedure of the routing table generation process by the layer 3 routing protocol processing unit. When the layer 3 routing protocol processing unit 17 receives the RIP response packet periodically transmitted from the adjacent node at the layer 3 level (step S101), the route entry in the packet is included in the routing table of the own node device. It is determined whether there is any (step S102). If the route entry is not included in the routing table of the own node device (Yes in step S102), an entry corresponding to the routing table is added (step S103), and a route information entry timer is set for the added entry. Start (step S104).

  If the route entry in the packet is included in the routing table of the own node device in Step S102 (No in Step S102), the next hop address of the RIP Response packet is the corresponding entry in the routing table of the own node device. It is determined whether or not the next hop information is different (step S105). When the next hop field value of the RIP Response packet is “0.0.0.0”, the next hop address is the IP address of the source node device, and the next hop field value is “0.0.0.0. If an address other than “0” is stored, the address is regarded as the next hop. If the next hop address of the RIP Response packet is different from the next hop information (Yes in step S105), is the metric of the route entry of the packet smaller than the metric of the corresponding entry in the routing table of the own node device? It is determined whether or not (step S106). When the metric of the route entry is smaller than the metric of the own node device (Yes in step S106), the next hop information of the corresponding entry in the routing table of the own node device is changed to the next hop address of the RIP Response packet. The metric value is also changed based on the route entry (step S107), and the route information entry timer for this entry is restarted (step S109). Further, when the metric of the route entry is equal to or more than the metric of the own node device (No in step S106), the corresponding entry is not updated.

  If the next hop information is the same in step S105 (No in step S105), the metric of the corresponding entry in the routing table of the own node device is set as the metric of the corresponding route entry in the RIP Response packet ( In step S108), the route information entry timer for this entry is restarted (step S109).

  After the above steps S104, S106, and S109, it is determined whether or not another route entry exists in the RIP Response packet (step S110). If another route entry exists (Yes in step S110). ) Again returns to step S102, and the above-described processing is repeatedly executed for other route entries. If there is no other route entry (No in step S110), the routing table generation process ends.

  If the route information entry timer expires without being restarted for a predetermined period, the route information (entry) is invalidated. Specifically, “16” is set in the metric of the entry, and the destination becomes unreachable. Here, the predetermined period (expiration time) generally set in the route information entry timer is 180 seconds.

  Further, in RIP, when a link of an invalidated interface is validated, the layer 3 routing protocol processing unit 17 may transmit a RIP Request packet to the interface. FIG. 22 is a diagram illustrating an example of the structure of a RIP Request packet. The RIP Request packet 400 includes a header part 401 and a route entry part 410. The header portion 401 includes RIP headers such as an IP header 402, a UDP header 403, an area 404 for storing a command field, and an area 405 for storing Version. The route entry unit 410 also includes an address family field 411, an IP address field 412 indicating the IP address of the destination network, a subnet mask field 413 indicating the subnet mask of the destination network, and a next hop (NextHop) indicating the next hop to the destination network. The field 414 includes a metric field 415 indicating a metric to the destination network.

  The node device on the same network that has received the RIP Request packet 400 returns the contents of the routing table of the own node device as a RIP Response packet. Here, in the received RIP Request packet 400, when the address family field 411 and the metric field 414 are set to “0” and all routes are designated, the adjacent node device at the layer 3 level Reply all of the contents of. When responding, information learned from the interface that transmits the RIP Response packet is not included in the response packet or the metric is “16”. Thus, by not including it in the response packet or setting the metric to “16”, the node device on the same network that has received the RIP Response packet does not learn the route information and prevents a route loop. It becomes possible. This procedure is widely known as Split Horizon technology. The Split Horizon technique is also used for RIP Response packets that are transmitted periodically. In the RIP Request packet 400, when the address family field 411 is set to “AF_INET”, that is, “2”, and a specific route is specified in the IP address field 412 and the subnet mask field 413, the layer 3 level The neighboring node device stores only the information of the designated route in the contents of the routing table in the RIP Response packet and responds. When a specific route is specified, a value on the routing table is set as a metric for information learned from the interface that transmits the RIP Response packet. In this case, Split Horizon is not performed.

  When the path information update unit 23 receives the topology information change detection notification from the topology change detection unit 20, the path information update unit 23 searches the ARP table 16a based on the physical address information included in the notification, and the node device corresponding to the physical address information The IP address is acquired, and the metric of the entry in the routing table in which the acquired IP address is set as the next hop is set to “16” to invalidate the route.

  The ARP table generation unit 15 transmits an ARP request to each node device on the layer 2 network upon receiving the RIP Response packet and recognizing the node device, and also transmits an ARP request every predetermined time. It has a function. In addition, the ARP table 16a is generated using the ARP response packet received from another node device.

  Here, an operation process performed by the network node device having such a configuration will be described. However, the network configuration is the same as that shown in FIG. 1 (FIG. 3) of the first embodiment. In addition, although the layer 3 routing protocol differs between the node device packet relay processing and the operation processing procedure of the node device adjacent to the node device that has failed in the layer 2 network 110 at the layer 2 level, the first embodiment is different. Therefore, the description thereof is omitted, and only the path information update process when an abnormality occurs in the node device in the layer 2 network 110 will be described.

  FIG. 23 is a flowchart showing the procedure of the route information update processing of the node device in the layer 2 network when a failure occurs. First, upon receiving a topology information packet transmitted from a node device adjacent to the failed node device at the layer 2 level (step S131), the node devices 100 to 104 update the topology map according to the contents of the topology information packet. As a result (step S132), the entry 504 for the node device 103 that has failed is removed from the topology map 14a. Next, the topology change detection unit 20 detects the node device that has disappeared due to the above-described update of the topology map (step S133), and sends a topology information change detection notification including the physical address information of the lost node device to the route information update unit 23. Notification is made (step S134).

  Upon receiving the topology information change detection notification, the path information update unit 23 searches the ARP table and acquires the IP address of the node device corresponding to the physical address information (step S135). Next, the route information update unit 23 searches the routing table 22a for an entry having the acquired IP address as the next hop, changes the metric of the entry to “16”, and invalidates the route indicated by the entry. (Step S136).

  Thereafter, the layer 3 routing protocol processing unit 17 receives the RIP response packet periodically transmitted by the adjacent node device on the layer 2 network (step S137), and information on the route entry included in the RIP response packet Then, the next hop and metric of the invalidated route are updated, and the route is validated (step S138). However, the disabled network is left in a disabled state for a destination network that cannot be reached. Thus, the update process of the route information of the node device when a failure occurs is completed.

  Next, regarding the specific packet relay processing of the node device at the time of the failure described above, the node device that relays the packet transmitted from the terminal A to the terminal B in the network configuration shown in FIG. 1 (FIG. 3) The relay process in 100 will be described as an example. However, the node device 100 is assumed to have the routing table 22a in FIG. 19, the ARP table 16a in FIG. 6, and the topology map 14a in FIG. 5 in a normal state (before a failure occurs).

  As shown in FIG. 14 from the state of FIG. 1, when a failure occurs in the node device 103 and stops, the node device 103 that is the next hop of the packet relayed from the node device 100 to the network N1 can be used for packet relay. Therefore, communication from terminal A to terminal B is stopped. At this time, the node device 102 adjacent to the node device 103 in which the failure has occurred at the layer 2 level detected the link failure of the second link connected to the node device 103, and displayed that the second link was a link failure. The topology information packet 116 is transmitted to the system 1 ring 112. Similarly, the node device 104 adjacent to the node device 103 in which the failure has occurred at the layer 2 level also detects the disconnection of the first link connected to the node device 103 and displays that the first link is a link failure. The topology information packet 115 is transmitted to the 0-system ring 111. These topology information packets 115 and 116 are relayed along the ring and reach all the node devices 100 to 102 and 104 on the layer 2 network 110 that is a ring network. It is assumed that the network N3 is also down due to the failure of the node device 103.

  Upon receiving the topology information packets 115 and 116, the layer 2 control unit 13 of the node device 100 updates the topology map according to the contents of the topology information packet, and as a result, the node device 103 that has failed from the topology map 14a. Entry 504 is removed. The topology map 14b generated by the node device 100 after the failure of the layer 2 network is the same as that shown in FIG. Next, the topology change detecting unit 20 of the node device 100 detects the lost node device 103 from the change in the topology map 14b shown in FIG. 15 from the topology map 14a in FIG. 5, and the physical address of the lost node device 103 is detected. The topology information change detection notification including the information “MAC3” is notified to the route information update unit 23. The path information update unit 23 searches the ARP table 16a of FIG. To get. Further, the route information update unit 23 searches the routing table 22a of FIG. 19 to extract entries 181 and 183 having the acquired IP address “10.0.0.103” as the next hop, and sets the metric to “ 16 ”to invalidate the route. FIG. 24 is a diagram illustrating an example of the contents of the routing table when a failure occurs. FIG. 24 shows a routing table 22b generated by the node device 100 as in FIG. As shown in FIG. 24, the metrics of the entries 721 and 723 corresponding to the destination network N1 and the network N3 having the IP address “10.0.0.103” of the node device 103 as the next hop are unreachable. The value shown is changed to “16”.

  Thereafter, the layer 3 routing protocol processing unit 17 of the node device 100 receives the RIP response packet periodically transmitted by the adjacent node device 102 at the layer 3 level. This RIP Response packet includes the contents of the routing table of the node device 102 as a path entry, and includes paths to the network N1 and the network N2. From the information of the route entry included in this packet, the IP address of the node device 102 that has transmitted the RIP Response packet is learned as the next hop to the destination network N1, and the corresponding entry in the routing table 22b is updated to validate the route. Turn into. FIG. 25 is a diagram illustrating an example of the contents of the routing table after the route is validated. FIG. 25 is also a routing table 22c generated by the node device 100 as in FIG. As shown in the routing table 22c, the next hop of the entry 721 corresponding to the destination address “network N1 (a pair of“ NetworkN1 ”and“ MaskN1 ””) is “10.0. And the metric is also changed from “16” to “3.” For an entry whose destination address is “network N3”, Split Horizon changes the route entry in the packet. Since it is not included, it remains disabled.

  After the next hop is changed and the routing table 22c is updated as shown in FIG. 25, the IP packet transmitted from the terminal A to the terminal B is a node whose node device 100 is a new next hop according to the routing table 22c. Since it is relayed to the device 102, the communication from the terminal A to the terminal B, which has been interrupted due to the occurrence of the failure of the node device 103, is resumed.

  According to the second embodiment, when a failure occurs in the layer 2 network 110 and the relay of the packet is stopped, it is possible to resume the communication after the maximum time of the RIP Response packet transmission interval elapses. For this reason, it is possible to significantly reduce the time until communication recovery. In other words, as compared with the conventional case of waiting until the expiration time of the route information entry timer in the routing table elapses and recognizing the occurrence of the failure, and then performing the route information change process and restarting communication, the time is reduced. This has the effect that it can be greatly shortened.

  Further, since the ARP table generation unit 15 manages the contents of the ARP table 16a so that there is always an entry of the ARP table 16a corresponding to the adjacent node device on the RPR network (layer 2 network 110), the topology information change When the detection notification is issued, the information (IP address) of the corresponding adjacent node device can always be acquired, and the path can be changed reliably.

Embodiment 3 FIG.
In the second embodiment, after the node device invalidates the corresponding entry in the routing table by the topology information change detection notification, the routing device is updated using the RIP Response packet periodically transmitted by the node device on the layer 2 network. In other words, the invalidated entry is validated, but in the third embodiment, the node device that invalidates the entry becomes a node device on the network when the corresponding entry in the routing table is invalidated. A case will be described in which a RIP Request packet for requesting route information is transmitted and a routing table is updated using the RIP Response packet obtained as a result.

  FIG. 26 is a block diagram schematically showing a functional configuration of the network node device according to the third embodiment of the present invention. When the network node devices 100 to 104 invalidate the entry in the routing table corresponding to the physical address information included in the topology information change detection notification by the route information update unit 23 in FIG. 18 of the second embodiment, The configuration further includes a route information inquiry unit 24 that transmits RIP request packets to adjacent node devices at the layer 3 level of the layer 2 network 110 to inquire route information. In addition, the same code | symbol is attached | subjected to the component same as FIG. 18 of Embodiment 2, and the detailed description is abbreviate | omitted.

  In the node device having such a configuration, the route information acquisition processing in the route information update processing when a failure occurs in the layer 2 network 110 will be described. The route information query unit 24 of the node device in FIG. An RIP request packet 400 for specifying all routes in which “request” is set in the command field 404 of the RIP header portion, “0” is set in the address family field 411 of the route entry portion 410, and “16” is set in the metric field 415. Send.

  On the other hand, the layer 3 routing protocol processing unit 17 of the node device on the layer 2 network 110 that has received the RIP request packet 400 transmits a RIP response packet storing the contents of the routing table of the own node device.

  Upon receiving the RIP Response packet, the Layer 3 routing protocol processing unit 17 of the node device that has transmitted the RIP Request packet 400 learns the route from the route entry included in the packet to the invalid destination network and updates the routing table. To do. That is, the source IP address of the node device that has transmitted the RIP Response packet is used as the “next hop” of the corresponding entry in the routing table, and the value of the metric field 415 of the route entry of the RIP Response packet is used as the “metric”. Use to update and validate the route to the invalidated destination network.

  Here, a specific example of the process of acquiring the path information of the node device when a failure occurs in the adjacent node device will be described. Here, as in the second embodiment, a description will be given of an operation process for updating route information of the node device 100 after a failure has occurred in the node device 103 in the network configuration shown in FIG. 1 (FIG. 3). As in the second embodiment, when the topology map 14a of the node device 100 is updated due to the failure of the node device 103, the topology change detection unit 20 sends a topology information change detection notification to the route information update unit 23. Be notified. When the route information update unit 23 invalidates an entry having the IP address corresponding to the physical address information included in the notification as the next hop, the route information query unit 24 uses the layer 2 network 110 (network N4). The RIP Request packet 400 shown in FIG. 22 is transmitted to the upper node device. In other words, as described above, the RIP Request packet 400 for specifying all routes in which “request” is set in the command field 404, “0” is set in the address family field 411, and the metric field 415 is set to “16” is transmitted. To do.

  The node device 102 that has received the RIP Request packet 400 transmits an RIP Response packet that stores the contents of the routing table of the own node device. However, the RIP Response packet 300 transmitted at this time includes route entries 310 and 320 whose destination IP addresses are routes to “network N1” and “network N2,” as shown in FIG. Split Horizon does not include a route entry whose destination IP address is a route to “network N3”, “network N4”, and “network N5”.

  When receiving the RIP response packet 300, the node device 100 learns the route from the route entry 310 included in the RIP response packet 300 to the “network N1”. That is, if the next hop field of the RIP Response packet 300 is “0.0.0.0”, the IP address of the source node device 102 included in the IP header 302 as the “next hop”, or the next hop field If the value is other than “0.0.0.0”, the value of the next hop field is used as “next hop”, and the value of the metric field 325 of the route entry 310 is used as “metric”. The entry 721 to be updated is updated, and the route to the “network N1” is validated. In this way, the routing table 22c shown in FIG. 25 is obtained. Since it is assumed that a failure has occurred in the network N3, the entry 723 in the routing table 22b remains invalidated.

  According to the third embodiment, immediately after the corresponding entry in the routing table due to the failure occurrence in the layer 2 network 110 is invalidated and the destination route is lost, the route information query unit 24 sends the RIP Request packet. By transmitting, it becomes possible to acquire information on alternative routes from adjacent node devices at the layer 3 level. Therefore, it has an effect that the time from the occurrence of a failure until the communication is restored can be significantly reduced as compared with the conventional case. Even if another node device that transmits the RIP Response packet does not implement the function of the third embodiment, the above-described effects can be obtained in the node device that implements the function of the third embodiment. it can. In other words, if the function of the third embodiment is implemented only in the node device at a necessary place on the layer 2 network 110, the cost is reduced without changing the node devices 100 to 104 of the entire layer 2 network 110. It becomes possible to construct a highly reliable network.

Embodiment 4 FIG.
In the third embodiment, when RIP is used as the layer 3 routing protocol, all the route information of adjacent node devices at the layer 3 level is invalidated when the corresponding entry in the routing table is invalidated by the topology information change detection notification. Although the case where a non-designated RIP Request packet is transmitted has been described, in the fourth embodiment, a case will be described in which only information related to the destination network of invalidated route information is requested by the RIP Request packet.

  FIG. 27 is a block diagram schematically showing a functional configuration of the network node device according to the fourth embodiment of the present invention. In FIG. 18 of the second embodiment, the network node devices 100 to 104 include a specific route information inquiry unit 26 that inquires specific layer information to an adjacent node device at the layer 3 level, and before the metric change of the routing table. A pre-update route information storage processing unit 27 for storing information, and a route information comparison unit 28 for comparing the route information before the update with the route information acquired as a result of the inquiry by the specific route information query unit 26 Have In addition, the same name and the same code | symbol are attached | subjected to the component which has the same function as FIG. 18 of Embodiment 2, and the description is abbreviate | omitted.

  The specific route information query unit 26 provides information on a specific route entry corresponding to the entry invalidated in the routing table by the route information update unit 23 to the node device on the layer 2 network 110 configured by RPR. Has a function of transmitting a RIP Request packet designated to transmit.

  The pre-update route information storage processing unit 27 receives the topology information change detection notification from the topology change detection unit 20, and when the route information update unit 23 invalidates the corresponding entry in the routing table, the pre-update route information ( (Hereinafter referred to as pre-update route information). More specifically, this pre-update route information is metric information of an entry to be invalidated in the routing table, and is stored as old metric information in the routing table.

  When receiving the RIP Response packet that is a response to the RIP Request packet transmitted by the adjacent node device at the layer 3 level, the route information comparing unit 28 receives the pre-update route information in the routing table, that is, the old metric of the invalidated entry. Compared with the information and the route entry included in the RIP Response packet corresponding to this entry, it has a function of determining whether or not the route entry in the RIP Response packet is adopted as new route information. Specifically, it is determined whether or not the metric information of the route entry of the RIP Response packet is smaller than the old metric information of the corresponding invalidated entry in the routing table, and if so, the route entry of the RIP Response packet is smaller. Is adopted as new route information, and is not adopted as route information in other cases.

  The routing table storage unit 19 is similar to the second embodiment in that it has a function of storing a routing table. In the fourth embodiment, the “old metric” added by the pre-update route information storage processing unit 27 is used. It differs from that of the second embodiment in that a routing table included as a new item is stored. FIG. 28 is a diagram illustrating an example of a routing table including pre-update route information. FIG. 28 shows the routing table 25a generated by the node device 100, as in FIG. 19 of the second embodiment. The item “old metric” stores metric information before the corresponding entry saved by the pre-update route information saving processing unit 27 is invalidated.

  Further, in addition to the function shown in the second embodiment, the layer 3 routing protocol processing unit 17 uses its own routing table as a response message to the RIP Request packet transmitted from the specific route information query unit 26 of another node device. It further has a function of setting and transmitting only the entry of a specific route stored in this packet in the RIP Response packet.

  Here, an operation process performed by the network node device having such a configuration will be described. However, the network configuration is the same as that shown in FIG. 1 (FIG. 3) of the first embodiment. In addition, although the layer 3 routing protocol differs between the node device packet relay processing and the operation processing procedure of the node device adjacent to the node device that has failed in the layer 2 network 110 at the layer 2 level, the first embodiment is different. Therefore, the description thereof is omitted, and only the path information update process when an abnormality occurs in the node device in the layer 2 network 110 will be described.

  FIG. 29 is a flowchart illustrating the procedure of the path information update process of the node device when a failure occurs in the layer 2 network. First, the same processing as steps S131 to S135 described in FIG. 23 of the second embodiment is performed, and processing for acquiring the IP address of the lost node device from the topology map 14a before update and the updated topology map 14b is performed. (Steps S151 to S155). That is, the layer 2 control unit 13 of the node device receives the topology information packet transmitted from the node device adjacent to the failed node device at the layer 2 level, and identifies the failed node device from the topology map 14a. When the topology map 14b is updated so as to be removed, the topology change detection unit 20 detects a node device that has disappeared due to the update of the topology map 14b described above, and sends a topology information change detection notification including physical address information of the lost node device. The route information update unit 23 is notified. Next, the path information update unit 23 that has received the topology change detection notification acquires the IP address of the node device corresponding to the physical address information included in the notification from the ARP table 16a.

  Thereafter, the pre-update route information storage processing unit 27 stores the metric information of the entry having the acquired IP address of the node device in the routing table as the next hop (step S156). Further, the route information update unit 23 invalidates the route of the entry having the acquired IP address in the routing table as the next hop (step S157). Next, the specific route information query unit 26 sets the route entry corresponding to the entry in the routing table in which the route is invalidated in step S157 as a specific route for the node device on the layer 2 network 110 to the route entry unit. The set RIP Request packet is transmitted (step S158).

  The adjacent node device at the layer 3 level that has received this RIP Request packet extracts the metric information corresponding to the specific route entry in the RIP Request packet from the routing table of the own node device, sets it in the RIP Response packet, and sets the RIP Response packet. The request packet is transmitted to the node device that transmitted the request packet.

  When this RIP Response packet is received (step S159), the path information comparison unit 28 of the node device compares the old metric of the invalidated entry in the routing table with the metric of the path entry included in the RIP Response packet. Then, it is determined whether or not a predetermined condition is satisfied (step S160). Here, examples of the predetermined condition include a case where the metric of the route entry included in the RIP Response packet is smaller than the value of the old metric of the invalidated entry.

  When the route information comparison unit 28 determines that the route entry in the RIP Response packet satisfies the predetermined condition (Yes in step S160), the route information comparison unit 28 newly sets the route entry included in the packet. The layer 3 routing protocol processor 17 updates the corresponding entry in the routing table (step S161). As a result, the route invalidated in step S157 is validated, and the processing of the node device at the time of occurrence of the failure ends. On the other hand, when the route information comparison unit 28 determines that the route entry in the RIP Response packet does not satisfy the predetermined condition (No in step S160), the route information comparison unit 28 includes the packet. The route entry is not adopted as new route information (step S162). That is, the invalid route in the routing table is left in that state. Thus, the route information update processing of the node device when a failure occurs in the layer 2 network 110 is completed.

  Next, regarding the specific packet relay processing of the node device at the time of the failure described above, the node device that relays the packet transmitted from the terminal A to the terminal B in the network configuration shown in FIG. 1 (FIG. 3) The relay process in 100 will be described as an example. However, it is assumed that the node device 100 has the routing table 25a in FIG. 28, the ARP table 16a in FIG. 6, and the topology map 14a in FIG. 5 in a normal state (before a failure occurs). Further, the node device 102 has learned the route to the network N3 from the node device 100, but does not implement the function of the fourth embodiment and has not detected the loss of the route due to the failure of the node device 103. And

  When the node device 103 of the layer 2 network 110 fails and stops from the state of FIG. 1 as shown in FIG. 14, the next hop of the packet relayed from the node device 100 to the network N1 (layer 2 network 110) Since the node device 103 is not usable for packet relay, the communication from the terminal A to the terminal B is stopped. At this time, the node device 102 adjacent to the node device 103 in which the failure has occurred at the layer 2 level detects the link failure of the second link connected to the node device 103 and displays that the second link is a link failure. The topology information packet is transmitted to the system 1 ring 112. Similarly, the node device 104 adjacent to the node device 103 in which the failure has occurred at the layer 2 level also detects the disconnection of the first link connected to the node device 103 and displays that the first link is a link failure. A topology information packet is transmitted to the 0-system ring 111. These topology information packets are relayed along the rings 111 and 112 and reach all the node devices 100 to 104 on the layer 2 network 110 which is a ring network.

  Upon receiving the topology information packets 115 and 116, the layer 2 control unit 13 of the node device 100 removes the entry for the failed node device 103 from the topology map 14a, and the connection relationship between the node devices after the failure has occurred. The topology map 14a is updated to reflect the above. The topology map 14b generated by the node device 100 after the failure of the layer 2 network 110 is the same as that shown in FIG. Next, the topology change detection unit 20 of the node device 100 detects the lost node device 103 from the changes in the topology maps 14a and 14b shown in FIGS. 5 to 15, and the physical address information “ The topology information change detection notification including “MAC3” is notified to the route information update unit 23. The path information update unit 23 searches the ARP table 16a of FIG. To get. At this time, the pre-update route information storage processing unit 27 extracts an entry having the IP address acquired by the route information update unit 23 as the next hop from the routing table, and stores the value of “metric” in “old metric”. Furthermore, the route information update unit 23 invalidates the route by setting the value of the “metric” to “16”. FIG. 30 is a diagram illustrating an example of the contents of the routing table when a failure occurs. FIG. 30 shows a routing table 25c generated by the node device 100 as in FIG. As shown in FIG. 30, the metrics of the entries 731 and 733 corresponding to the destination addresses “network N1” and “network N3” having the IP address “10.0.0.103” of the node device 103 as the next hop are shown. Is changed to a value “16” indicating unreachable, and “old metric” is inputted with metric values “3” and “2” before being invalidated.

  Thereafter, the specific route information inquiry unit 26 of the node device 100 transmits an RIP Request packet in which the invalidated entry is set as a specific route in the route entry unit to the adjacent node device at the layer 3 level. FIG. 31 is a diagram illustrating an example of the contents of the RIP Request packet transmitted by the specific route information inquiry unit. The configuration of the RIP Request packet 400a is the same as that described with reference to FIG. 22, except that “request” is set in the command field 404 of the header section 401, and the route entry sections 420 and 430 indicate routes in the routing table. A path entry corresponding to the invalidated entries 731 and 733 is stored. In other words, the route entry 420 is set with the route having the destination address “network N1”, and the route entry 430 is set with the route having the destination address “network N3”. At this time, “2” is set in the address family field 421.

  The adjacent node device 102 at the layer 3 level that has received the RIP request packet 400a uses only the entry in the routing table of the node device 102 corresponding to the specific route specified by the RIP request packet 400a as the route entry unit. The set RIP response packet is transmitted. FIG. 32 is a diagram showing an example of the contents of this RIP Response packet. The configuration of the RIP response packet 300a is the same as that described with reference to FIG. 20. In this example, the route of the “network N1” is set as the destination network in the route entry 330 as a response to the RIP request packet 400a of FIG. In the route entry 340, the route of “network N3” is set as the destination network.

  The node device 100 receives the RIP response packet 300a from the node device 102, and the route information comparison unit 28 sets the “destination” of the entry 731 in which the IP address field of the route entry 330 included in this packet is invalidated in the routing table 25b. Since the metric value “2” is smaller than the stored old metric value “3”, the route entry 330 is adopted as new route information. As a result, the layer 3 routing protocol processing unit 17 stores the IP address “10.0.0.102” of the node device 102 in the next hop of the routing table entry 731, and the metric of the received route entry 260 as the metric. “3” which is a value obtained by adding 1 to is stored and the entry is validated. Also, for the route entry 340 of the received packet, since the metric value “2” is equal to the stored old metric value “2” of the corresponding entry 733 in the routing table 25b, it is not adopted as new route information. . That is, the entry 733 remains invalidated. FIG. 33 is a diagram illustrating an example of the contents of the updated routing table. FIG. 33 is also a routing table 25c generated by the node device 100, as in FIG. As shown in the routing table 25c, the next hop and metric values corresponding to the entry 731 are validated from the invalidated state of FIG.

  When another adjacent node device at the layer 3 level, such as the node device 104, also receives the RIP Request message transmitted from the node device 100, the RIP Response message that stores information on the destination addresses “network N1” and “network N3”. Send. However, the route entry included in the RIP Response message transmitted from the node device 104 is learned by the RIP Response message transmitted from the node device 102 or the node device 103, and the value of the metric is the node device 102 or the node device 103. 1 is added to the metric value in the RIP Response message transmitted by Therefore, in the node device 100 that has received the RIP Response message from the node device 104, the stored old metric value is equal to the metric value of the received route entry, so that the route information transmitted from the node device 104 by the node device 100. It is possible to prevent loops in the path.

  According to the fourth embodiment, the node device that has detected the change in the topology information requests only the information related to the destination network of the invalidated route information from the adjacent node device at the layer 3 level. Since only the information (route information) about the designated invalidated route is responded, there is an effect of reducing the load on the layer 2 network 110. The larger the network scale and the more routes out of the ring, the more effective.

As described above, the network node device and the route information update method thereof according to the present invention are suitable when a layer 2 network is configured with an RPR network and the node device has a layer 3 routing protocol processing function.

It is a figure which shows typically an example of the network structure using a ring-shaped layer 2 network. It is a block diagram which shows the function structure of Embodiment 1 of the network node apparatus by this invention. FIG. 2 is a logical configuration diagram of a network when the network configuration of FIG. 1 is viewed at a layer 3 level. 2 is a table showing addresses, identification information, and the like set for each node in the network of FIG. 1. It is a figure which shows an example of a structure of a topology map. It is a figure which shows an example of an ARP table. It is a figure which shows typically the content of an OSPF Hello packet. It is a flowchart which shows the procedure of the production | generation process of a neighbor registration table. It is a flowchart which shows the procedure of the close examination process of a neighbor registration table. It is a figure which shows an example of the content of a neighbor registration table. It is a figure which shows an example of a data structure of a routing table. It is a flowchart which shows the procedure of the operation | movement process of the packet relay by the node apparatus when a layer 2 network is normal. It is a flowchart which shows the procedure of the operation | movement process of the node apparatus adjacent to the node apparatus in which the failure in a layer 2 network generate | occur | produced. It is a flowchart which shows the procedure of the operation process of the node apparatus in a layer 2 network at the time of a failure generation. FIG. 2 is a diagram illustrating an example of a network configuration when an abnormality occurs in a layer 2 network in FIG. 1. It is a figure which shows an example of the topology map produced | generated after abnormality occurrence of a layer 2 network. It is a figure which shows an example of the neighbor registration table from which the entry was deleted. It is a figure which shows an example of the recalculated routing table. It is a block diagram which shows the function structure of Embodiment 2 of the network node apparatus by this invention. It is a figure which shows an example of a routing table. It is a figure which shows an example of the structure of a RIP Response packet. It is a flowchart which shows the procedure of the production | generation process of a routing table. It is a figure which shows an example of the structure of a RIP Request packet. It is a flowchart which shows the procedure of the operation process of the node apparatus in the layer 2 network at the time of a failure occurrence. It is a figure which shows an example of the content of the routing table at the time of failure occurrence. It is a figure which shows an example of the content of the routing table after enabling a path | route. It is a block diagram which shows typically the function structure of Embodiment 3 of the network node apparatus by this invention. It is a block diagram which shows typically the function structure of Embodiment 4 of the network node apparatus by this invention. It is a figure which shows an example of the routing table containing the route information before update. It is a flowchart which shows the procedure of the operation process of the node apparatus in a layer 2 network at the time of a failure generation. It is a figure which shows an example of the content of the routing table at the time of failure occurrence. It is a figure which shows an example of the content of the RIP Request packet transmitted by the specific route information inquiry part. It is a figure which shows an example of the content of this RIP Response packet. It is a figure which shows an example of the content of the updated routing table.

Explanation of symbols

11 First link control unit 12 Second link control unit 13 Layer 2 control unit 14 Topology map storage unit 15 ARP table generation unit 16 ARP table storage unit 17 Layer 3 routing protocol processing unit 18 Neighbor registration table storage units 19, 22, 25 Routing table storage unit 20 Topology change detection unit 21 Connection information update unit 23 Route information update unit 24 Route information inquiry unit 26 Specific route information inquiry unit 27 Pre-update route information storage processing unit 28 Route information comparison units 100 to 104 Network node Device (node device)
105 Layer 3 node device 111 0 system ring 112 1 system ring

Claims (10)

A network node device that constitutes a network node device of a layer 2 network and performs data relay processing according to a layer 3 routing protocol,
When the link state changes in addition to the predetermined time interval, the topology information including the link state with the node device adjacent to the own node device is transmitted and based on the topology information received from the other node device. Layer 2 control means for generating topology map information indicating a connection state of the layer 2 network;
For a node device in the layer 2 network, a logical address used in layer 3 corresponding to the physical address is acquired from a physical address used in layer 2, and an ARP table that associates the physical address with the logical address is generated. ARP table generation means;
A layer 3 routing protocol processing means for generating a routing table based on the layer 3 routing protocol and performing data relay processing according to the routing table;
A neighbor registration table storage means for storing a neighbor registration table including a node device adjacent to the node device on the layer 3 network;
Topology change detection means for detecting a change in the topology map information and notifying a connection information update means of a topology information change detection notification including a physical address of a node device lost from the topology map information ;
Connection information update means for acquiring a logical address corresponding to the physical address of the topology information change detection notification from the ARP table and deleting an entry in the neighbor registration table having the acquired logical address ;
With
The network node device, wherein the layer 3 routing protocol processing means regenerates a routing table when detecting an update of the neighbor registration table. A network node device that constitutes a network node device of a layer 2 network and performs data relay processing according to a layer 3 routing protocol,
When the link state changes in addition to the predetermined time interval, the topology information including the link state with the node device adjacent to the own node device is transmitted and based on the topology information received from the other node device. Layer 2 control means for generating topology map information indicating a connection state of the layer 2 network;
For a node device in the layer 2 network, a logical address used in layer 3 corresponding to the physical address is acquired from a physical address used in layer 2, and an ARP table that associates the physical address with the logical address is generated. ARP table generation means;
Look including the following relay destination for relaying the received data to the destination network, and the next hop is the next relay destination for relaying to the destination network of the received data, the metric is the number of hops to the destination network A routing table having an entry , and periodically exchanging route information including the contents of the routing table between adjacent node devices at the layer 3 level to generate a routing table, and relaying data based on the routing table Layer 3 routing protocol processing means for processing;
Topology change detection means for detecting a change in the topology map information and notifying a path information update means of a topology information change detection notification including a physical address of a node device lost from the topology map information ;
Path information update means for acquiring a logical address corresponding to the physical address of the topology information change detection notification from the ARP table, and invalidating an entry in the routing table having the acquired logical address as a next hop ;
Route information inquiry means for requesting transmission of route information to an adjacent node device at the layer 3 level when an entry is invalidated by the route information update means;
Equipped with a,
The Layer 3 routing protocol processing means, the network characterized by features and to Rukoto to update the routing table based on the contents of the route information received from the adjacent node device layer 3 level after the invalidation of the entry Node device. The network node device according to claim 2 , wherein the route information inquiry unit further includes a function of requesting an adjacent node device at the layer 3 level to transmit only the route information for the invalidated entry. Pre-update route information storage processing means for storing a metric of the entry as an old metric when an entry in the routing table is invalidated by the connection information update means;
Specific route information inquiry means for requesting the adjacent node device of the layer 3 level only to transmit route information for the invalidated entry when the entry is invalidated by the route information update means;
Whether the metric corresponding to the invalidated entry in the route information obtained by the specific route information inquiry means satisfies a predetermined condition with respect to the old metric saved by the pre-update route information saving processing means Route information comparison means for determining whether or not,
Further comprising
The Layer 3 routing protocol processing unit, when by the path information comparison means in said path information satisfies the predetermined condition, and updates the routing table using the routing information according to claim 2 or 4. The network node device according to 3 . The ARP table generating means, to the node device of the layer 2 network, any one of claims 2 to 4, characterized in that it has a function of inquiring the correspondence between logical addresses and physical addresses periodically The network node device described in 1. A method of updating path information of a network node device that constitutes a network node device of a layer 2 network and performs data relay processing according to a layer 3 routing protocol,
A topology map information update step of updating topology map information indicating a connection state of the layer 2 network based on topology information transmitted when a link state between node devices of the layer 2 network changes;
When a change in the topology map information is detected, a logical address used in the layer 3 network corresponding to a physical address used in the layer 2 network of the deleted entry in the topology map information is changed to a node in the layer 2 network. The device acquires from the ARP table in which the physical address used in layer 2 and the logical address used in layer 3 are associated with the device , and an entry including this logical address is stored in the local node device on the layer 3 network. A neighbor registration table update step for removing from a neighbor registration table including a node device adjacent to
When detecting the update of the neighbor registration table, a routing table regeneration step for regenerating the routing table;
A path information update method for a network node device, comprising: A route information updating method for a network node device that constitutes a network node device of a layer 2 network and performs data relay processing according to a layer 3 routing protocol,
Next hop as a next relay destination for periodically exchanging route information including the contents of the routing table of the own node device with the adjacent node device and relaying to the destination network of the received data A routing table generating step for generating a routing table having as an entry a metric that is the number of hops to the destination network ;
A topology map information update step of updating topology map information indicating a connection state of the layer 2 network based on topology information transmitted when a link state between node devices of the layer 2 network changes;
When a change in the topology map information is detected, a logical address used in the layer 3 network corresponding to a physical address used in the layer 2 network of the deleted entry in the topology map information is acquired, and this logical address is obtained. An entry invalidation step of invalidating an entry in the routing table as a next hop ;
A route information inquiry step for requesting transmission of route information to an adjacent node device at the layer 3 level when an entry is invalidated in the entry invalidation step;
As a result of the route information query step, a routing table update step for updating the routing table based on the contents of the route information received from the adjacent node device at the layer 3 level after invalidation of the entry ;
A path information update method for a network node device, comprising: 8. The route information updating method for a network node device according to claim 7 , wherein, in the route information inquiry step, the adjacent node device at the layer 3 level is requested only to transmit the route information for the invalidated entry. . When an entry in the routing table is invalidated in the routing table update step, a process of storing a metric of the entry as an old metric is further performed.
A specific route information query step for requesting an adjacent node device at a layer 3 level only to transmit route information for the invalidated entry when an entry is invalidated in the routing table update step;
When the metric corresponding to the invalidated entry in the route information obtained by the specific route information query step satisfies a predetermined condition with respect to the stored old metric, the route information is used. A route information comparison step for updating the routing table;
The route information update method for a network node device according to claim 7 or 8 , further comprising: The network node device according to any one of claims 7 to 9 , further comprising an address association step of periodically associating a physical address and a logical address of the network node device in the layer 2 network. Route information update method.

Images