Communication control system, communication control program, and communication control method

Abstract

To make it possible to maintain communication even in the case that a failure of hardware / software of a communication channel or a layer 2 switch has occurred.SOLUTION: In a communication control system according to the invention: each of groups includes a plurality of communication controllers; each communication controller includes a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit; in each group, the first switch units form a tree structure and second switch units form a tree structure; a most-significant first switch unit and a most-significant second switch unit of the tree structures in the group can perform inter-group communication; the first switch unit and the second switch unit of a least-significant communication controller in the group can perform communication; and upon detecting a failure, the control unit of each communication controller switches connection between the first switch unit and the second switch unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a communication control system, a communication control program, and a communication control method. [Background technology]

[0002] There is a demand for route redundancy in the event of a failure in a Layer 2 network.

[0003] 2 and 3 are configuration diagrams showing a network configuration of a voice communication unit (communication control device) that provides an IP voice communication service within a company. Fig. 2 and Fig. 3 show a single-system network configuration that is not duplicated.

[0004] 2 and 3, one base rack unit (#0) 91-0 and five extension rack units (#1 to #5) 91-1 to 91-5 are distinguished, but basically they are all servers with the same functions. The base rack unit (#0) 91-0 and the extension rack units (#1 to #5) 91-1 to 91-5 each form a layer 2 network by controlling a layer 2 switch (L2SW) by a control unit 901.

[0005] Here, a Layer 2 switch (L2SW) performs data transfer functions at the second layer (data link layer) of the OSI reference model. An example of a data transfer device is a repeater hub, which broadcasts the same data to all ports, but a Layer 2 switch (L2SW) has a destination assigned to each port, and transfers data only to the destination port.

[0006] In the single-system Layer 2 networks of Figures 2 and 3, the L2SW forwards data to the destination port of the packet, so if the L2SW fails or there is a fault in the communication line, communication between L2SWs is not possible and services such as voice communication are stopped.

[0007] To duplicate a Layer 2 network, the bonding function is used as a conventional function (see Figure 4).

[0008] As shown in the example in Figure 4, bonding is a function that bundles two physical ETHERNET (registered trademark; the same applies below) into one logical port to improve load balancing and fault tolerance. In the bonding mechanism, the physical ETHERNET is called the slave interface, and one logical port is called the master interface.

[0009] Bonding determines which of the two layer 2 network ETHERNETs, ​​"Eth#0" and "Eth#1," to use and switches between them.

[0010] Bonding has the following seven functions to determine when to switch between modes (hereafter referred to as "modes").

[0011] "1.mode=0 balance-rr" This is load balancing using round robin. In other words, it switches between ETHERNET in turn. This is not suitable for fault tolerance.

[0012] "2.mode=1 active-backup" Two ETHERNET ports are divided into "active" and "backup", communication is performed through one active slave interface, and when the active port fails, it switches to the backup slave interface. This is suitable for failure countermeasures.

[0013] "3.mode=2 balance-xor" Load balancing is done based on source / destination information. This is not suitable for fault tolerance.

[0014] "4.mode=3 broadcast" The same packets are sent from all slave interfaces, which is not suitable for fault tolerance.

[0015] "5.mode=4 802.3ab" Load balancing is performed based on source and destination information. LACP (Link Aggregation Control Protocol) is used for reception. This is not suitable for fault tolerance.

[0016] "6.mode=5 balance-tlb" The load is balanced according to the load of the slave interface and transmission is switched. This mode is not supported on some Linux distributions, so please be careful whether it is supported or not.

[0017] "7.mode=6 balance-alb" The load is balanced according to the load of the slave interface and transmission is switched. Reception is also switched according to the load. This mode is not supported on some Linux distributions, so please be careful whether it is supported or not. [Prior art documents] [Non-patent literature]

[0018] [Non-Patent Document 1] Hidematsu Kasano, "Linux High Reliability Server Construction Guide: Single Server Edition", CQ Publishing, May 15, 2009, pp. 58-59 Summary of the Invention [Problem to be solved by the invention]

[0019] In the case of the network configurations shown in Figures 2 and 3, if a hardware or software failure occurs, such as a break or disconnection of the communication line or L2SW, a communication failure will occur immediately. The mechanism is not an issue but a specification, but in an environment where high maintainability is required, a countermeasure that can maintain communication is desirable.

[0020] As described above, even if a hardware configuration for Ethernet duplication is implemented and software is equipped with bonding in order to withstand failures in Layer 2 networks such as LANs (registered trademark), the following problems still remain.

[0021] There are two methods for detecting a fault by the control unit 901: link-up fault detection and detection based on whether an arp response is possible.

[0022] Problems with link-up failure detection will be explained with reference to Fig. 5. As shown in Fig. 5(A), communication is performed via "Eth#0". As shown in Fig. 5(B), when a failure occurs in L2SW#01 that is directly connected to "Eth#0", the uplink of "Eth#0" goes down, so the control unit 910 can detect the failure. Therefore, the control unit 910 switches the connection to "Eth#1" to continue communication. However, when a failure occurs in L2SW#02 (the L2SW connected beyond L2SW#01) that is not directly connected to "Eth#0", as shown in Fig. 5(C), the control unit 910 cannot detect the failure, so there is a problem that ETHERNET cannot be switched.

[0023] On the other hand, when detection is performed using an arp response, the control unit 910 is able to detect a fault because it is no longer able to detect an arp response to an arp request, as described below, and is therefore able to switch to ETHERNET. This will be explained with reference to FIG.

[0024] As shown in Fig. 6(A), communication is performed via "Eth#0". As shown in Fig. 6(B), when a failure occurs in L2SW#01 that is directly connected to "Eth#0", the arp response to the arp request is not transmitted, so the control unit 910 can detect the failure. Therefore, the control unit 910 switches the connection to "Eth#1" to continue communication. Also, when a failure occurs in L2SW#02 (the L2SW connected beyond L2SW#01) that is not directly connected to "Eth#0", as shown in Fig. 6(C), the arp response to the arp request is not transmitted, so the control unit 910 can detect the failure and switch to ETHERNET.

[0025] However, there are problems such as unnecessary ARP-related packets flowing within the broadcast domain, overhead due to ARP monitoring, and the need to review the filtering settings of switch functions.

[0026] In addition, although it is possible to set multiple IP addresses to wait for ARP responses, if there is a response from any of them, it will operate as if there is no problem, which means that it is not possible to switch ETHERNET.

[0027] It is possible to set a condition that there is no problem if responses are received from all IP addresses for which ARP responses are being waited for; however, this makes IP management complicated and makes maintenance difficult.

[0028] Therefore, there is a demand for a communication control system, a communication control program, and a communication control method that can avoid communication, maintain communication, and extend the life of communication processing even when a hardware or software failure occurs in a layer 2 network, such as damage to a communication line or layer 2 switch, or a cable becoming unplugged. [Means for solving the problem]

[0029] In order to solve such problems, the first invention provides a communication control system in which each of a plurality of groups has a plurality of communication control devices and performs redundant control of communication between the respective communication control devices, characterized in that (1) each communication control device has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit, (2) for each group, each of the multiple first switch units belonging to the first network forms a tree structure, and each of the multiple second switches belonging to the second network forms a tree structure, (3) the first switch unit and the second switch unit of the communication control device located at the top position of the tree structure within the group are capable of communication between the groups, (4) the first switch unit and the second switch unit of the communication control device located at the bottom position of the tree structure within the group are capable of communication, and (5) the control unit of each communication control device switches the connection between the first switch unit and the second switch unit when a fault is detected.

[0030] A second aspect of the present invention is a communication control system in which each of a plurality of groups includes a plurality of communication control devices, each of which has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit, and for each group, each of the multiple first switch units belonging to the first network form a tree structure, and each of the multiple second switches belonging to the second network form a tree structure, and the first switch unit and the second switch unit of the communication control device located at the top position of the tree structure within the group are capable of communicating between groups, and the first switch unit and the second switch unit of the communication control device located at the bottom position of the tree structure within the group are capable of communicating, characterized in that the computer of the communication control device is made to function as a control unit that switches the connection between the first switch unit and the second switch unit when a fault is detected.

[0031] A third aspect of the present invention is a communication control system in which each of a plurality of groups includes a plurality of communication control devices, each of which has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit, and for each group, each of the multiple first switch units belonging to the first network form a tree structure, and each of the multiple second switches belonging to the second network form a tree structure, and the first switch unit and the second switch unit of the communication control device located at the top position of the tree structure within the group are capable of communication between groups, and the first switch unit and the second switch unit of the communication control device located at the bottom position of the tree structure within the group are capable of communication, characterized in that the control unit of the communication control device switches the connection between the first switch unit and the second switch unit when a fault is detected. Effect of the Invention

[0032] According to the present invention, even when a hardware or software failure occurs in a layer 2 network, such as damage to a communication line or layer 2 switch, or a cable being pulled out, communication can be avoided, communication can be maintained, and the life of communication processing can be extended. [Brief description of the drawings]

[0033] [Figure 1] FIG. 1 is a configuration diagram showing a network configuration according to an embodiment. [Diagram 2] FIG. 1 is a diagram showing a conventional network configuration (part 1). [Diagram 3] FIG. 2 is a diagram showing a conventional network configuration (part 2). [Figure 4] FIG. 1 is an explanatory diagram illustrating a bonding function in a conventional unit. [Diagram 5] FIG. 1 is an explanatory diagram for explaining a problem of redundancy in a conventional network (part 1). [Figure 6] FIG. 2 is an explanatory diagram for explaining a problem with redundancy in a conventional network (part 2). [Figure 7]FIG. 2 is a configuration diagram showing a connection configuration between L2SWs in a network according to an embodiment. [Figure 8] FIG. 1 is an explanatory diagram for explaining processing in a network according to an embodiment (part 1); [Figure 9] FIG. 2 is an explanatory diagram for explaining processing in a network according to an embodiment (part 2); [Figure 10] FIG. 11 is an explanatory diagram for explaining processing in a network according to an embodiment (part 3). [Figure 11] FIG. 4 is an explanatory diagram for explaining processing in a network according to an embodiment (part 4); [Figure 12] FIG. 5 is an explanatory diagram for explaining processing in a network according to an embodiment; [Figure 13] FIG. 6 is an explanatory diagram for explaining processing in a network according to an embodiment; [Figure 14] FIG. 7 is an explanatory diagram for explaining processing in a network according to an embodiment; [Figure 15] FIG. 8 is an explanatory diagram for explaining processing in a network according to an embodiment; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] (A) Main embodiment DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a communication control system, a communication control program, and a communication control method according to the present invention will be described in detail with reference to the drawings.

[0035] (A-1) Configuration of the embodiment (A-1-1) Network configuration FIG. 1 is a configuration diagram showing a network configuration according to an embodiment.

[0036] In FIG. 1, a network 100 as a communication control system has a basic rack unit (#0) 10-0 as a communication control device and a plurality of extension rack units (#1 to #5) 10-1 to 10-5.

[0037] The basic rack unit (#0) 10-0 and the extension rack units (#1 to #5) 10-1 to 10-5 basically perform the same processing.

[0038] For example, the basic rack unit (#0) 10-0 and the extension rack units (#1-#5) 10-1-10-5 are voice communication devices (voice communication units) that control voice communication such as internal lines and external lines within a company, and the network 100 can be a voice communication network that controls internal lines and external lines within a company. Note that, although a voice communication network is illustrated here, it is not limited to voice communication as long as it is a Layer 2 network.

[0039] The basic frame unit (#0) 10-0 functions as a master device in the network 100, and has a predetermined number of ports related to voice communication services.

[0040] The extension rack units (#1 to #5) 10-1 to 10-5 are installed in accordance with the excess number of ports accommodated when a number of ports exceeding the number accommodated by the basic rack unit (#0) 10-0 is required.

[0041] In other words, functionally, the basic rack unit 10-0 and the extension rack units (#1 to #5) 10-1 to 10-5 are the same. In this example, five extension rack units (#1 to #5) 10-1 to 10-5 are added as extension racks. The number of extension rack units to be added is not particularly limited.

[0042] Each of the base rack unit (#0) 10-0 and the extension rack units (#1 to #5) 10-1 to 10-5 has two layer 2 switches (L2SW), and the L2SWs of each unit are connected to each other. In other words, the network 100 constitutes a layer 2 network that transfers data at the second layer (layer 2) of the OSI reference model.

[0043] Moreover, the base rack unit (#0) 10-0 and the extension rack units (#1 to #5) 10-1 to 10-5 each include a layer 2 switch (L2SW) belonging to Eth#0 and a layer 2 switch (L2SW) belonging to Eth#1, forming a duplicated layer 2 network. A detailed description of the duplicated configuration will be given later.

[0044] The network 100 is roughly divided into two logical groups, group G1 and group G2. These groups are also called unit groups.

[0045] The group G1 includes a basic rack unit (#0) 10-0, an extension rack unit (#1) 10-1, and an extension rack unit (#2) 10-2.

[0046] The group G2 includes the extension rack unit (#3) 10-3, the extension rack unit (#4) 10-4, and the extension rack unit (#5) 10-2.

[0047] The number of logical groups is not limited to 2 and may be 3 or more. Also, the number of units belonging to one group is not particularly limited as long as there are more than one unit.

[0048] The basic frame unit (#0) 10-0 has a layer 2 switch (L2SW#00) that connects to one of the duplicated layer 2 networks (ETHERNET), "Eth#0", a layer 2 switch (L2SW#01) that connects to the other layer 2 network, "Eth#1", and a control unit 101.

[0049] For example, in the notation "#01" of "L2SW#01", the second digit "0" indicates the unit number, and the first digit "1" indicates the type of layer 2 network "Eth#1" to which the layer 2 switch is connected. The same rules are used to notate other layer 2 switches.

[0050] Each of the expansion rack units (#1 to #5) 10-1 to 10-5, like the basic rack unit (#0) 10-0, has a layer 2 switch (L2SW) that connects to "Eth#0" on one side, a layer 2 switch (L2SW) that connects to "Eth#1" on the other side, and a control unit 101.

[0051] The control unit 101 is responsible for the functions of the basic rack unit (#0) 10-0 and the extension rack units (#1 to #5) 10-1 to 10-5. The control unit 101 includes, for example, a CPU, a ROM, a RAM, an EEPROM, etc., and the CPU executes a processing program (for example, a communication control program that performs duplex control) to realize the functions of the unit.

[0052] The control unit 101 can use an operating system (OS) such as Linux (registered trademark), and has a failure detection unit 110 and a bonding unit 120 as functions installed in the OS.

[0053] The fault detection unit 110 detects a communication fault in the network 100. There are various events that can cause a communication fault, but in this embodiment, it does not matter what the cause of the fault is. Communication faults between L2SWs include a LAN cable being pulled out or broken, and also include equipment failures (hardware or software failures) of the L2SW. In any case, the fault detection unit 110 can widely apply existing technologies, for example, a link-up fault detection method can be applied. Note that the link-up fault detection method can apply "mode=1 active-backup" of bonding as the bonding unit 120.

[0054] The bonding unit 120 bundles a plurality of network interfaces together to provide redundancy. The bonding unit 120 is sometimes called network teaming, and for example, bonding installed in Linux can be applied.

[0055] (A-1-2) Network duplication configuration 7 is a configuration diagram showing a connection configuration between L2SWs in a network 100 according to an embodiment. The connection configuration between L2SWs shown in FIG. 7 is the same as the connection configuration between L2SWs in FIG.

[0056] In the base rack unit (#0) 10-0, the L2SW #00 is connected to the L2SW #10 and is also connected to the L2SW #30 of the extension rack unit (#3) 10-3 belonging to group G2. The L2SW #01 is connected to the L2SW #11 and is also connected to the L2SW #31 of the extension rack unit (#3) 10-3 belonging to group G2.

[0057] In the extension rack unit (#1) 10-1, the L2SW#10 is connected to the L2SW#00 and is also connected to the L2SW#20. The L2SW#11 is connected to the L2SW#01 and is also connected to the L2SW#21.

[0058] In the expansion rack unit (#2) 10-2, the L2SW #20 is connected to the L2SW #10 and is also connected to the other layer 2 switch, L2SW #21, in the expansion rack unit (#2) 10-2. The L2SW #21 is also connected to the L2SW #20 in the expansion rack unit (#2) 10-2 and is also connected to the L2SW #11.

[0059] In the extension rack unit (#3) 10-3, the L2SW #30 is connected to the L2SW #00 of the base rack unit (#0) 10-0 of the group G1, and is also connected to the L2SW #40. The L2SW #31 is connected to the L2SW #01 of the base rack unit (#0) 10-0 of the group G1, and is also connected to the L2SW #41.

[0060] In the extension rack unit (#4) 10-4, the L2SW #40 is connected to the L2SW #30 and also to the L2SW #50. Also, the L2SW #41 is connected to the L2SW #31 and also to the L2SW #51.

[0061] In the expansion rack unit (#5) 10-5, the L2SW #50 is connected to the L2SW #40 and is also connected to the other layer 2 switch, L2SW #51, in the expansion rack unit 10-5. The L2SW #51 is also connected to the L2SW #50 in the expansion rack unit (#5) 10-5 and is also connected to the L2SW #41.

[0062] By connecting the L2SWs in this manner, "L2SW#00", "L2SW#10", "L2SW#20", "L2SW#30", "L2SW#40" and "L2SW#50" constitute "Eth#0".

[0063] Furthermore, "L2SW#01", "L2SW#11", "L2SW#21", "L2SW#31", "L2SW#41" and "L2SW#51" constitute "Eth#1".

[0064] Here, the configuration of the layer 2 network within the groups (group G1, group G2) will be described.

[0065] For example, in "Eth#0" of group G1, if L2SW#00 of basic rack unit (#0) 10-0 is the "parent node," then L2SW#10 of expansion rack unit (#1) 10-1 is the "child node" of L2SW#00, and L2SW#20 of expansion rack unit (#2) 10-2 is the "child node" of L2SW#10, forming a tree structure in Eth#0 of group G1.

[0066] In "Eth#1" of group G1, if L2SW#01 of basic rack unit (#0) 10-0 is the "parent node," L2SW#11 of expansion rack unit (#1) 10-1 is the "child node" of L2SW#01, and L2SW#21 of expansion rack unit (#2) 10-2 is the "child node" of L2SW#11, forming a tree structure in Eth#1 of group G1.

[0067] In group G1, in the expansion rack unit (#2) 10-2 that is located at the position furthest from the parent node (the end position of the tree structure within the group), two layer 2 switches, L2SW #20 and L2SW #21, are connected.

[0068] Similarly, in group G2, a tree structure is formed for each Eth#0 (or Eth#1), and further, L2SW#50 and L2SW#51 that are located at the ends of group G2 are connected.

[0069] In this way, by using the connection configuration illustrated in FIG. 7, the network 100 can form a loop configuration in which Eth#0 and Eth#1 are connected.

[0070] Furthermore, each L2SW of the basic unit (#0) 10-0 and the expansion units (#1 to #5) 10-1 to 10-5 implements the Rapid Spanning Tree Protocol (RSTP) to construct a spanning tree configuration between the L2SWs.

[0071] In this way, by implementing RSTP in L2SWs, the convergence time for state consistency between L2SWs when a failure occurs can be shortened, enabling rapid route avoidance.

[0072] (A-2) Operation of the embodiment Next, the operation of the duplex processing in the network 100 according to this embodiment will be described in detail with reference to the drawings.

[0073] In the following, for convenience of explanation, when expressing a basic rack unit (#0) 10-0, for example, it will be expressed as "basic rack unit #0" and the like.

[0074] (A-2-1) Normal communication processing 8 is an explanatory diagram (part 1) for explaining the processing in the network 100 according to the embodiment. In Fig. 8, thick solid lines indicate communication routes, and thick dotted lines indicate control routes.

[0075] 8, the L2SW configuration of network 100 is a loop configuration. In other words, the connections are as follows: "L2SW#00" - "L2SW#10" - "L2SW#20" - "L2SW#21" - "L2SW#11" - "L2SW#01" - "L2SW#31" - "L2SW#41" - "L2SW#51" - "L2SW#50" - "L2SW#40" - "L2SW#30" - "L2SW#00" -...

[0076] Of the two ETHERNETs, ​​"Eth#0" is in the active state and "Eth#1" is in the standby state (inactive state). In other words, "Eth#0" is the normal communication route.

[0077] In group G1, data is transferred between "L2SW#00", "L2SW#10" and "L2SW#20" which belong to "Eth#0", and in group G2, data is transferred between "L2SW#30", "L2SW#40" and "L2SW#50" which belong to "Eth#0".

[0078] Furthermore, data is transferred between groups between "L2SW#00" of group G1 and "L2SW#30" of group G2, which belong to "Eth#0."

[0079] At this time, in the base rack unit 10-0 and each of the extension rack units (#1 to #5) 10-1 to 10-5, the failure detection unit 110 of the control unit 101 monitors the link state of the network interface connected to the active L2SW.

[0080] Therefore, when the failure detection unit 110 detects that a failure has occurred in the layer 2 switch (L2SW) belonging to "Eth#0", the bonding unit 120 switches the connection of the layer 2 switch (L2SW) to implement path avoidance. This will be explained below.

[0081] (A-2-2) A failure occurred on L2SW#20 FIG. 9 is an explanatory diagram (part 2) for explaining the processing in the network 100 according to the embodiment.

[0082] Consider the case where some kind of failure occurs in L2SW #20 of expansion unit #2 while packets are being transferred via the normal communication route.

[0083] First, a failure occurs in the L2SW #20 in the expansion unit #2 (S101).

[0084] The fault detection unit 110 of the control unit 101 of the extension rack unit #2 monitors the network interface connected to the active L2SW #20, and detects a link-up down, thereby detecting a fault in the L2SW #20 (S102).

[0085] When a failure in the L2SW #20 is detected, the bonding unit 120 of the control unit 101 switches the connection to the L2SW #21 in a standby (inactive) state (S103).

[0086] Then, L2SW#21 notifies L2SW#51 of extension unit#5 of information regarding the implementation of connection switching by RSTP (S104). For example, L2SW#21 notifies L2SW#51 of an instruction to change the connection. In response to this, L2SW#51 notifies L2SW#21 that the connection switching to L2SW#51 is to be enabled, and L2SW#21 returns a response to this to L2SW#51, thereby implementing the connection switching.

[0087] In the extension unit rack #5, the control unit 101 performs port assignment so that data destined for L2SW #21 passes through L2SW #50 and L2SW #51.

[0088] Then, in the extension rack unit #5, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S105). As a result, data addressed to L2SW #21 is provided from L2SW #50 to L2SW #51 (S106), and further, the data is provided from L2SW #51 to L2SW #21 via "Eth #1" which was in the standby state.

[0089] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (basic rack unit 10-0, extension rack unit 10-1, and extension rack units 10-3 to 10-5) can communicate without switching the L2SW.

[0090] (A-2-3) A failure occurs on L2SW#10 FIG. 10 is an explanatory diagram for explaining the processing in the network 100 according to the embodiment (part 3).

[0091] Consider the case where some kind of failure occurs in L2SW #10 of expansion unit #1 while packets are being transferred via the normal communication route.

[0092] First, in the expansion rack unit #1, a failure occurs in the L2SW #10 (S201).

[0093] The fault detection unit 110 of the control unit 101 of the extension rack unit #1 monitors the network interface connected to the active L2SW #10, and detects a link-up down, thereby detecting a fault in the L2SW #10 (S202).

[0094] When a failure in the L2SW #10 is detected, the bonding unit 120 of the control unit 101 switches the connection to the L2SW #11 in a standby (inactive) state (S203).

[0095] Then, the L2SW #11 notifies the L2SW #51 in the extension rack unit #5 and the L2SW #21 in the extension rack unit #2 of information related to the implementation of connection switching by RSTP (S204).

[0096] For example, L2SW #11 issues a connection change instruction to each of L2SW #51 and L2SW #21.

[0097] In response to this, L2SW #51 notifies L2SW #11 that the connection switching to the L2SW #51 is to be enabled, and L2SW #11 returns a response to this to L2SW #51, thereby implementing the connection switching.

[0098] Similarly, L2SW #21 also notifies L2SW #11 to enable connection switching to L2SW #21, and L2SW #11 returns a response to this to L2SW #21, thereby performing connection switching.

[0099] As a result, in the extension rack unit #5 of group G2, the control unit 101 performs port allocation so that data addressed to L2SW #21 passes through L2SW #50 and L2SW #51.

[0100] Then, in the extension rack unit #5, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S205). As a result, data addressed to L2SW #11 is provided from L2SW #50 to L2SW #51 (S206), and further, the data is provided from L2SW #51 to L2SW #11 via "Eth #1" which was in the standby state.

[0101] Similarly, in the extension rack unit #2 of group G1, the control unit 101 performs port assignment so that data addressed to L2SW #11 passes through L2SW #21 and L2SW #20.

[0102] Then, in the extension rack unit #2, the control unit 101 enables the connection between L2SW #21 and L2SW #20 (S207). As a result, data addressed to L2SW #11 is provided from L2SW #21 to L2SW #20 (S208), and further, the data is provided from L2SW #21 to L2SW #20 via "Eth #1" which was in the standby state.

[0103] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (basic rack unit 10-0, extension rack unit 10-2, and extension rack units 10-3 to 10-5) can communicate without switching the L2SW.

[0104] (A-2-4) A failure occurred on L2SW#00 FIG. 11 is an explanatory diagram for explaining the processing in the network 100 according to the embodiment (part 4).

[0105] Consider the case where some kind of failure occurs in L2SW #00 of basic unit #0 while packets are being transferred via the normal communication route.

[0106] First, a failure occurs in L2SW #00 in basic unit #0 (S301).

[0107] The fault detection unit 110 of the control unit 101 of the basic frame unit #0 monitors the network interface connected to the active L2SW #00, and detects a link-up down, thereby detecting a fault in the L2SW #00 (S302).

[0108] When a failure in L2SW#00 is detected, the bonding unit 120 of the control unit 101 switches the connection to L2SW#01 in a standby (inactive) state (S303).

[0109] Then, the L2SW #11 notifies the L2SW #51 in the extension rack unit #5 and the L2SW #21 in the extension rack unit #2 of information related to the implementation of connection switching by RSTP (S304).

[0110] For example, L2SW #01 issues a connection change instruction to each of L2SW #51 and L2SW #21.

[0111] In response to this, L2SW#51 notifies L2SW#01 that the connection switching to the L2SW#51 is to be enabled, and L2SW#01 returns a response to this to L2SW#51, thereby implementing the connection switching.

[0112] Similarly, L2SW #21 also notifies L2SW #01 that the connection switching to L2SW #21 is to be enabled, and L2SW #01 returns a response to this to L2SW #51, thereby implementing the connection switching.

[0113] As a result, in the extension rack unit #5 of group G2, the control unit 101 performs port assignment so that data addressed to L2SW #01 passes through L2SW #50 and L2SW #51.

[0114] Then, in the extension rack unit #5, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S305). As a result, data addressed to L2SW #01 is provided from L2SW #50 to L2SW #51 (S306), and further, the data is provided from L2SW #51 to L2SW #11 via "Eth #1" which was in the standby state.

[0115] Similarly, in the extension rack unit #2 of group G1, the control unit 101 assigns ports so that data addressed to L2SW #01 passes through L2SW #21-L2SW #20.

[0116] Then, in the extension rack unit #2, the control unit 101 enables the connection between L2SW #21 and L2SW #20 (S307). As a result, data addressed to L2SW #01 is provided from L2SW #21 to L2SW #20 (S308), and further, the data is provided from L2SW #21 to L2SW #20 via "Eth #1" which was in the standby state.

[0117] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (extension rack units 10-1 to 10-2, extension rack units 10-3 to 10-5) can communicate without switching the L2SW.

[0118] (A-2-5) A failure occurred on L2SW#30 FIG. 12 is an explanatory diagram for explaining the processing in the network 100 according to the embodiment (part 5).

[0119] Consider the case where some kind of failure occurs in L2SW #30 of expansion unit #3 while packets are being transferred via the normal communication route.

[0120] First, in the expansion rack unit #3, a failure occurs in the L2SW #30 (S401).

[0121] The fault detection unit 110 of the control unit 101 of the extension rack unit #3 monitors the network interface connected to the active L2SW #30, and detects a link-up down, thereby detecting a fault in the L2SW #30 (S402).

[0122] When a failure in the L2SW #30 is detected, the bonding unit 120 of the control unit 101 switches the connection to the L2SW #31 in a standby (inactive) state (S403).

[0123] Then, the L2SW #31 notifies the L2SW #51 in the expansion rack unit #5 and the L2SW #21 in the expansion rack unit #2 of information related to the implementation of connection switching by RSTP (S404).

[0124] For example, L2SW #31 issues a connection change instruction to each of L2SW #51 and L2SW #21.

[0125] In response to this, L2SW#51 notifies L2SW#31 that the connection switching to L2SW#51 is to be enabled, and L2SW#31 returns a response to this to L2SW#51, thereby implementing the connection switching.

[0126] Similarly, L2SW #21 also notifies L2SW #31 to enable the switching of the connection to L2SW #21, and L2SW #31 returns a response to this to L2SW #21, thereby performing the switching of the connection.

[0127] As a result, in the extension rack unit #5 of group G2, the control unit 101 performs port allocation so that data addressed to L2SW #31 passes through L2SW #50 and L2SW #51.

[0128] Then, in the extension rack unit #5, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S405). As a result, data addressed to L2SW #31 is provided from L2SW #50 to L2SW #51 (S406), and further, the data is provided from L2SW #51 to L2SW #31 via "Eth #1" which was in the standby state.

[0129] Similarly, in the extension rack unit #2 of group G1, the control unit 101 performs port assignment so that data addressed to L2SW #31 passes through L2SW #21 and L2SW #20.

[0130] Then, in the extension rack unit #2, the control unit 101 enables the connection between L2SW #21 and L2SW #20 (S407). As a result, data addressed to L2SW #31 is provided from L2SW #21 to L2SW #20 (S408), and further, the data is provided from L2SW #21 to L2SW #20 via "Eth #1" which was in the standby state.

[0131] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (basic rack unit 10-0, extension rack units 10-1 to 10-2, extension rack units 10-4 to 10-5) can communicate without switching the L2SW.

[0132] (A-2-6) A failure occurred on L2SW#40 FIG. 13 is an explanatory diagram (part 6) for explaining the processing in the network 100 according to the embodiment.

[0133] Let us consider the case where some kind of failure occurs in L2SW #40 of expansion unit #4 while packets are being transferred via the normal communication route.

[0134] First, a failure occurs in L2SW#40 in expansion unit#4 (S501).

[0135] The fault detection unit 110 of the control unit 101 of the extension rack unit #4 monitors the network interface connected to the active L2SW #40, and detects a link-up down, thereby detecting a fault in the L2SW #40 (S502).

[0136] When a failure in the L2SW #40 is detected, the bonding unit 120 of the control unit 101 switches the connection to the L2SW #41 in a standby (inactive) state (S503).

[0137] Then, the L2SW #41 notifies the L2SW #51 in the expansion rack unit #5 and the L2SW #21 in the expansion rack unit #2 of information related to the implementation of connection switching by RSTP (S504).

[0138] For example, L2SW #41 issues a connection change instruction to each of L2SW #51 and L2SW #21.

[0139] In response to this, L2SW #51 notifies L2SW #41 that the connection switching to the L2SW #51 is to be enabled, and L2SW #41 returns a response to this to L2SW #51, thereby implementing the connection switching.

[0140] Similarly, L2SW #21 also notifies L2SW #41 that the connection switching to L2SW #21 is to be enabled, and L2SW #41 returns a response to this to L2SW #21, thereby implementing the connection switching.

[0141] As a result, in the extension rack unit #5 of group G2, the control unit 101 performs port assignment so that data addressed to L2SW #41 passes through L2SW #50 and L2SW #51.

[0142] Then, in the extension rack unit #5, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S505). As a result, data addressed to L2SW #41 is provided from L2SW #50 to L2SW #51 (S506), and further, the data is provided from L2SW #51 to L2SW #41 via "Eth #1" which was in the standby state.

[0143] Similarly, in the extension rack unit #2 of group G1, the control unit 101 performs port assignment so that data addressed to L2SW #41 passes through L2SW #21 and L2SW #20.

[0144] Then, in the extension rack unit #2, the control unit 101 enables the connection between L2SW #21 and L2SW #20 (S507). As a result, data addressed to L2SW #41 is provided from L2SW #21 to L2SW #20 (S508), and further, the data is provided from L2SW #21 to L2SW #20 via "Eth #1" which was in the standby state.

[0145] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (basic rack unit 10-0, extension rack units 10-1 to 10-2, extension rack units 10-3, 10-5) can communicate without switching the L2SW.

[0146] (A-2-7) A failure occurred on L2SW#50 FIG. 14 is an explanatory diagram (part 7) for explaining the processing in the network 100 according to the embodiment.

[0147] Let us consider the case where some kind of failure occurs in L2SW #50 of expansion unit #5 while packets are being transferred via the normal communication route.

[0148] First, a failure occurs in L2SW #50 in expansion unit #5 (S601).

[0149] The fault detection unit 110 of the control unit 101 of the extension rack unit #5 monitors the network interface connected to the active L2SW #50, and detects a link-up down, thereby detecting a fault in the L2SW #50 (S602).

[0150] When a failure in the L2SW #50 is detected, the bonding unit 120 of the control unit 101 switches the connection to the L2SW #51 in a standby (inactive) state (S603).

[0151] Then, L2SW #51 notifies L2SW #21 of extension unit #5 of information relating to the implementation of connection switching by RSTP (S604).

[0152] For example, L2SW #51 issues a connection change instruction to L2SW #21.

[0153] In response to this, L2SW #21 notifies L2SW #51 that the connection switching to L2SW #21 is to be enabled, and L2SW #51 returns a response to this to L2SW #21, thereby implementing the connection switching.

[0154] As a result, in the extension rack unit #2 of group G1, the control unit 101 performs port allocation so that data addressed to L2SW #51 passes through L2SW #21 and L2SW #20.

[0155] Then, in the extension rack unit #2, the control unit 101 enables the connection between L2SW #21 and L2SW #20 (S605). As a result, data addressed to L2SW #51 is provided from L2SW #21 to L2SW #20 (S606), and further, the data is provided from L2SW #21 to L2SW #20 via "Eth #1" which was in the standby state.

[0156] In this way, in the network 100, the layer 2 network is no longer in a loop configuration, but the other units (basic rack unit 10-0, extension rack units 10-1 to 10-2, extension rack units 10-3 to 10-4) can communicate without switching the L2SW.

[0157] (A-2-8) Fault occurs between L2SW FIG. 15 is an explanatory diagram (part 7) for explaining the processing in the network 100 according to the embodiment.

[0158] Consider the case where some kind of failure occurs in the communication line between L2SW #00 and L2SW #30 while packets are being forwarded via the normal communication route.

[0159] A failure occurs in the communication line between L2SW #00 of group G1 and L2SW #30 of group G2 (S701).

[0160] If a fault occurs in the communication line, the control unit 101 cannot detect the fault, and therefore data cannot be transferred using the communication line between L2SW #00 of group G1 and L2SW #30 of group G2.

[0161] However, when transmission data occurs in the basic rack unit #0 or any of the extension rack units #1 to #5, in the extension rack unit #5 of group G2, the control unit 101 enables the connection between L2SW #50 and L2SW #51 (S702) and starts up L2SW #51 (S703).

[0162] Furthermore, in the extension rack unit #2 of the group G1, the control unit 101 enables the connection between the L2SW #20 and the L2SW #21 (S704), and starts up the L2SW #21 (S705).

[0163] By doing this, data can be transferred through "Eth#1", which was in standby state.

[0164] (A-3) Effects of the embodiment As described above, according to this embodiment, even when a hardware or software failure occurs such as damage to a communication line or layer 2 switch in a layer 2 network, or a cable becoming unplugged, communication can be avoided, communication can be maintained, and the life of communication processing can be extended.

[0165] (B) Other embodiments Although various modified embodiments have been mentioned in the above-described embodiment, the present invention can also be applied to the following modified embodiments.

[0166] (B-1) In the above embodiment, when the present invention is applied to an audio communication system including an audio communication device, the L2SW of the audio communication device has a duplex configuration. However, the present invention can also be applied to a duplex configuration in which each device interface has a duplex configuration in a network including other system devices.

[0167] (B-2) In the above embodiment, the simplest tree structure is illustrated for each ETHERNET in the group, but the tree structure may have a "branch" structure in which edges connecting nodes are arranged. In other words, the tree structure may have branches between nodes. [Explanation of symbols]

[0168] 100...network, 10-0...basic rack unit (#0), 10-1 to 10-5...extension rack units (#1 to #5), 101...control unit, 110...fault detection unit, 120...bonding unit.

Claims

1. In a communication control system in which each of a plurality of groups includes a plurality of communication control devices and performs redundant control of communication between the respective communication control devices, Each of the communication control devices has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit; For each of the groups, a plurality of the first switches belonging to the first network each form a tree structure, and a plurality of the second switches belonging to the second network each form a tree structure, the first switch unit and the second switch unit of the communication control device present at the highest position of a tree structure in the group are capable of communicating between groups, the first switch unit and the second switch unit of the communication control device that is present at the lowest position of a tree structure in the group are capable of communicating with each other, The control unit of each of the communication control devices switches the connection between the first switch unit and the second switch unit when a fault is detected. A communication control system comprising:

2. the first switch unit and the second switch unit of the communication control device that is present at the lowest position of the tree structure are connected to each other for each group; The first network and the second network form a loop configuration.

2. The communication control system according to claim 1 .

3. The communication control system according to claim 1, characterized in that the control unit of each of the communication control devices treats a port of the first switch unit and a port of the second switch unit as a single logical port, and switches the connection between the first switch unit and the second switch unit when a fault is detected.

4. Each of the plurality of groups includes a plurality of communication control devices; Each of the communication control devices has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit; For each of the groups, a plurality of the first switches belonging to the first network each form a tree structure, and a plurality of the second switches belonging to the second network each form a tree structure, the first switch unit and the second switch unit of the communication control device present at the highest position of a tree structure in the group are capable of communicating between groups, The first switch unit and the second switch unit of the communication control device that is present at the lowest position of the tree structure in the group are capable of communicating with each other. In a communication control system, A computer of the communication control device, a control unit that switches the connection between the first switch unit and the second switch unit when a fault is detected; A communication control program that causes the computer to function as a

5. Each of the plurality of groups includes a plurality of communication control devices; Each of the communication control devices has a first switch unit belonging to a first network, a second switch unit belonging to a second network, and a control unit; For each of the groups, a plurality of the first switches belonging to the first network each form a tree structure, and a plurality of the second switches belonging to the second network each form a tree structure, the first switch unit and the second switch unit of the communication control device present at the highest position of a tree structure in the group are capable of communicating between groups, The first switch unit and the second switch unit of the communication control device that is present at the lowest position of the tree structure in the group are capable of communicating with each other. In a communication control system, A control unit of the communication control device When a fault is detected, the connection between the first switch unit and the second switch unit is switched. A communication control method comprising:

Images