WO2013014764A1

WO2013014764A1 - Communication system, communication apparatus and communication method

Info

Publication number: WO2013014764A1
Application number: PCT/JP2011/067099
Authority: WO
Inventors: 浩司柴田; 健志北山
Original assignee: 三菱電機株式会社
Priority date: 2011-07-27
Filing date: 2011-07-27
Publication date: 2013-01-31

Abstract

Provided is a communication network including a plurality of ring networks each of which comprises a plurality of nodes, uses two adjacent nodes as ring connection nodes and is physically connected to another ring network via these ring connection nodes. Each ring network communicates with an adjacent ring network via one of the ring connection nodes, and if the ring connection node in communication with the adjacent ring network encounters concurrent communication failures with both of the two nodes adjacent thereto within the local ring network, the ring network notifies the adjacent ring network of a two-path failure to that effect, and further switches the ring connection node communicating with the adjacent ring network. If the ring network receives a notification of a two-path failure from the adjacent ring network, the ring network switches the ring connection node communicating with the adjacent ring network.

Description

COMMUNICATION SYSTEM, COMMUNICATION DEVICE, AND COMMUNICATION METHOD

The present invention relates to a communication system formed by a plurality of ring networks.

In a ring network, a single port on the ring network is blocked to prevent a loop from occurring (the same signal continues to be transferred in the ring network). In addition, there has been proposed a method of switching communication paths so as to bypass a failure when a failure occurs in the ring network. At this time, in order to avoid the occurrence of a loop, the port to be blocked is switched as appropriate.

For example, in ERP (Ethernet Ring Protection), which is a protection standard for Ethernet (registered trademark) rings, a single port is blocked in the master node, thereby preventing a normal loop from occurring. When a failure occurs, the node that detected the failure closes the port on which the failure has occurred, and transmits a control frame for failure occurrence notification. When the master node that has received the control frame releases the port block, the path switching is completed (see Non-Patent Document 1).

In a multi-ring network in which ring networks are connected, it is necessary to make the nodes connecting the ring networks redundant from the viewpoint of fault tolerance. That is, the ring networks are physically connected using a plurality of nodes, and the ports are blocked at some of these nodes (called ring connection nodes). For example, when two ring connection nodes are physically connected, the port is blocked on one side. However, since the above ERP is defined assuming a single ring, the following problems occur when applied to a multi-ring network.

Consider a multi-ring network configured as shown in FIG. In FIG. 15, a ring network A (hereinafter referred to as ring A) is formed of nodes a1 to a5, and a ring network B (hereinafter referred to as ring B) is formed of nodes b1 to b5. In ring A, node a4 closes the port to avoid a loop, and in ring B, node b4 closes the port to avoid the loop. Of the nodes a1, a2, b1, and b2, which are ring connection nodes, the nodes a2 and b2 block the ports, and signals (frames) are not transmitted and received on the path (link) between the nodes a2 and b2. Is set to Signal transmission / reception between ring A and ring B is performed via nodes a1 and b1.

In the multi-ring network having such a configuration, when a failure as shown in FIG. 16, that is, a failure occurs in a path (link) on both sides of the node a1 that performs signal transmission / reception with another ring, In ring A, the port blocking setting is changed by ERP, and the port blocking in the node a4 is released (in the nodes a2 and a5, the port on the fault occurrence side (node a1 side) is blocked) . However, since the path connecting the rings (the link between the nodes a1 and b1, the link between the nodes a2 and b2) is not subject to ERP, the port is blocked in the path connecting the rings. The setting is not changed. In the examples of FIGS. 15 and 16, the port blockage at the nodes a2 and b2 is not released. As a result, communication between ring A and ring B becomes impossible. That is, in the ring network, when a failure occurs on the path on both sides of a node that is a connection point with another ring network, there is a problem that communication between the rings becomes impossible.

The present invention has been made in view of the above, and in a multi-ring network, even when a failure occurs in a path on both sides of a node that is a connection point with another ring network, communication between the ring networks is performed. An object of the present invention is to obtain a communication system, a communication apparatus, and a communication method that can continue the communication.

In order to solve the above-described problems and achieve the object, the present invention includes a plurality of ring networks including a plurality of nodes, and each ring network includes two adjacent nodes as ring connection nodes. A communication system that is physically connected to another ring network via a connection node, wherein each ring network communicates with an adjacent ring network via one of the ring connection nodes, and the adjacent ring network When a communication failure occurs between the ring connection node in communication with both of the two adjacent nodes in the own ring network at the same time, notification of the occurrence of both path failures to that effect to the adjacent ring network and Switch the ring connection node to communicate with the ring network Even when receiving the notification of both pathways failure from that the ring network, and wherein the switching ring connection node in communication with the adjacent ring network.

In the communication system according to the present invention, when the communication between the rings becomes impossible, it is possible to recognize the change and change the port blocking setting. Even if it does, there exists an effect that communication between rings can be continued.

FIG. 1 is a diagram showing a configuration example of a first embodiment of a communication system according to the present invention. FIG. 2 is a diagram illustrating a configuration example of a ring connection node. FIG. 3 is a flowchart illustrating an operation example of the inter-ring failure notification transmission / reception unit. FIG. 4 is a flowchart illustrating an operation example of the inter-ring failure detection unit. FIG. 5 is a flowchart illustrating an operation example of the redundant pair node failure determination unit. FIG. 6 is a flowchart illustrating an operation example of the both-path failure detection unit. FIG. 7 is a flowchart illustrating an operation example of the inter-ring path switching control unit. FIG. 8 is a diagram illustrating a block connection port block control operation performed by the inter-ring path switching control unit. FIG. 9 is a diagram illustrating a control operation when both path failures occur. FIG. 10 is a diagram illustrating a control operation when a failure occurs in the inter-ring route. FIG. 11 is a diagram illustrating a control operation when a ring connection node fails. FIG. 12 is a diagram illustrating a configuration example of a ring connection node according to the second embodiment. FIG. 13 is a diagram illustrating a block connection port block control operation performed by the inter-ring path switching control unit. FIG. 14 is a diagram illustrating a control operation when a failure that has occurred in an inter-ring path is recovered. FIG. 15 is a diagram for explaining the problem. FIG. 16 is a diagram for explaining the problem.

Hereinafter, embodiments of a communication system, a communication device, and a communication method according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
FIG. 1 is a diagram showing a configuration example of a first embodiment of a communication system according to the present invention. The communication system of the present embodiment includes a ring network (hereinafter referred to as a ring) A formed by nodes a1 to a5 and a ring B formed by nodes b1 to b5. Ring A and ring B are connected to each other by node a1 and node b1, node a2 and node b2. Note that the number of nodes forming the rings A and B is not limited to that shown in FIG. In order to simplify the explanation, the number of rings is 2, but it may be 3 or more.

In rings A and B, ports are blocked in some nodes in order to avoid the occurrence of loops, and signals are not transferred beyond the blocked ports. In the example of FIG. 1, the node a4 closes the port on the node a3 side, and the node b4 closes the port on the node b5 side. Each node periodically transmits a control signal (failure detection control signal) for detecting the occurrence of a communication failure to each adjacent node (adjacent node). If a failure detection control signal cannot be received from an adjacent node for a certain period of time, it is determined that a communication failure has occurred, the port on which the failure has occurred is blocked, and a control signal indicating failure occurrence (failure notification) Control signal) is transmitted within the own ring (transmitted from the non-blocked port toward the adjacent node). When each node receives the failure notification control signal, the node releases the blocking if the port on the side different from the receiving side is blocked. If the port on the side different from the receiving side is not blocked, the failure notification control signal is transferred to an adjacent node on the side different from the receiving side. In each ring, for example, by applying ERP, each node detects a communication failure and notifies other nodes. In each of the embodiments including the present embodiment, description will be made assuming that ERP is applied. Note that other control having the same function as ERP may be applied.

Among the nodes connecting the rings, one node is set as a redundant node, and the redundant node is set not to transmit / receive signals to / from another ring. In the example of FIG. 1, the nodes a2 and b2 are redundant nodes, and these nodes block ports on the opposite node side (connection ports with other rings).

In the following description, for convenience of description, each node (nodes a1, a2, b1, b2) connecting the rings is referred to as a ring connection node. A pair of ring connection nodes on the same ring (a pair of nodes a1 and a2, a pair of nodes b1 and b2) is called a redundant pair node. A communication path (also called a link) through which ring connection nodes of different rings communicate is called an inter-ring connection link. As shown in FIG. 1, the node a1 and the node b1 are physically connected by an inter-ring connection link 10-1, and the node a2 and the node b2 are physically connected by an inter-ring connection link 10-2.

Among the ring connection nodes, a node that does not block a connection port with another ring, that is, a node that communicates with a ring connection node of another ring (referred to as an opposite ring connection node) (node a1 in the example of FIG. 1). And b1) periodically transmits a control signal (hereinafter referred to as a multi-ring protection control frame) for monitoring the state of the communication path to other rings. When a multi-ring protection control frame cannot be received from another ring for a certain period, it is determined that a communication failure has occurred, and the port connecting the inter-ring connection link where the communication failure has occurred is blocked. Notify other ring connection nodes of the same ring that a communication failure has occurred in the inter-ring connection link. When this notification is received, the ring connection node releases the block of the port to which the inter-ring connection link is connected. In this way, the ring connection node monitors the occurrence of a communication failure on the inter-ring connection link, and if a failure is detected, it notifies the other redundant pair node and releases the port blockage. Even if a failure occurs in the link, communication between rings can be maintained by changing the port blocking setting.

FIG. 2 is a diagram showing a configuration example of a ring connection node corresponding to the communication apparatus according to the present invention, and shows a configuration example of a node a1 that is a ring connection node. The configuration of the other ring connection nodes (nodes a2, b1, b2) is the same as that of the node a1.

As shown in FIG. 2, the node a1 includes a PHY unit 11, a ring connection port I / F (Interface) unit 12, a frame multiplexing control unit 13 having a buffer control unit 14, a west port I / F unit 15, East port I / F unit 16, PHY units 17 and 18, single ring protection unit 19 (hereinafter referred to as SRP unit 19), multi-ring protection unit 20 (hereinafter referred to as MRP unit 20), , FDB unit 21, ring connection port 22, West ring port 23, and East ring port 24.

The SRP unit 19 includes an intra-ring fault detection unit 191, an intra-ring fault notification reception unit 192, an intra-ring protection unit 193, and an intra-ring fault notification transmission unit 194. The ring A includes adjacent nodes (nodes a2, a2). The communication state with a5) is monitored to detect a communication failure (including a failure detection notification operation to other nodes) and perform port blocking control (blocking setting / cancellation).

The MRP unit 20 includes an inter-ring failure notification transmission / reception unit 201, an inter-ring failure detection unit 202, a redundant pair node failure determination unit 203, a both-path failure detection unit 204, and an inter-ring route switching control unit 205, Monitors the communication status with (node b1) to detect a communication failure (including failure detection notification operation to other ring connection nodes), and block the port used for communication with the opposite ring connection node (Set / release blocking).

The configuration of the remaining nodes (nodes a3, a4, a5, b3, b4, and b5) other than the ring connection node is obtained by deleting the MRP unit 20 and the components and functions related thereto from the ring connection node. Note that the same configuration as that of the ring connection node may be used and the function of the MRP unit 20 may not be used.

The PHY unit 11 transmits and receives signals to and from other ring nodes (ring connection nodes) via the ring connection port 22. In reception processing, frame data is extracted from a received signal that is a communication medium signal arriving from the inter-ring connection link 10-1 (in the case of nodes a2 and b2, the inter-ring connection link 10-2) via the ring connection port 22. Then, it is transferred to the ring connection port I / F unit 12 in the form of a frame. In the transmission process, the frame data received from the ring connection port I / F unit 12 is converted into a communication medium signal and transmitted from the ring connection port 22.

The ring connection port I / F unit 12 transmits / receives a frame to / from the PHY unit 11. In the reception process, first, the validity of the received frame is confirmed. If it is not valid (there is an error), the frame is discarded. If it is valid, the type of the received frame is confirmed, and address learning information is extracted from the received frame and output to the FDB unit 21. In addition, the destination of the received frame is notified to the FDB unit 21 to perform an address search, and based on the destination address search result obtained from the FDB unit 21, a port to be sent (West ring port 23 or East ring port) 24) is selected. If the selected port is not blocked, the frame is transferred to the frame multiplexing control unit 13 together with the selected port information (selected port information). If the received frame is a multi-ring protection control frame as a result of the type confirmation, the control information included in the received frame is output to the inter-ring failure notification transmission / reception unit 201, and the arrival of the multi-ring protection control frame is confirmed. The arrival information shown is output to the inter-ring failure detection unit 202.

Also, the ring connection port I / F unit 12 outputs the multi-ring protection control frame received from the inter-ring failure notification transmission / reception unit 201 to the PHY unit 11 in the transmission process. Further, when a frame is received from the frame multiplexing control unit 13 while the ring connection port 22 is not blocked, this frame is output to the PHY unit 11.

The frame multiplexing control unit 13 individually performs frame multiplexing control (Add / Drop / Transit) including buffer control by the buffer control unit 14 for each of the West ring port 23 and the East ring port 24. In other words, although not shown in the drawing, it is composed of a functional unit that performs frame multiplexing control on the West ring port 23 and a functional unit that performs frame multiplexing control on the East ring port 24. The frame multiplexing control unit 13 includes the frame of the Add traffic input from the ring connection port I / F unit 12 and the frame of the Transit traffic transferred from the West port I / F unit 15 or the East port I / F unit 16. Two-input one-output transmission arbitration is performed for multiplexing and outputting to the ring. The frame multiplexing control unit 13 multiplexes the drop traffic frame input from the West port I / F unit 15 or the East port I / F unit 16 and outputs the frame to the ring connection port 22. Perform output transmission arbitration. When the frame multiplexing control unit 13 receives a frame from the ring connection port I / F unit 12, the West port I / F unit 15 or the East port I / F unit 16, the port indicated by the selected port information sent together with the frame The frame is output to the port I / F unit (ring connection port I / F unit 12, West port I / F unit 15 or East port I / F unit 16) corresponding to.

Further, when the buffer control unit 14 of the frame multiplexing control unit 13 receives a buffer clear instruction from the inter-ring path switching control unit 205, the frame stored in the buffer is discarded.

The West port I / F unit 15 transmits and receives frames to and from the PHY unit 17. In the reception process, first, the validity of the received frame is confirmed. If it is not valid (there is an error), the frame is discarded. If it is appropriate, the type of received frame is confirmed. In the case of an ERP control frame (control frame used for route setting in the ring, failure detection, etc.), arrival information indicating the arrival of the ERP frame is detected as failure in the ring. The control information included in the ERP frame is output to the in-ring failure notification receiving unit 192. Also, information for address learning is extracted from the received frame and output to the FDB unit 21. Further, the destination of the received frame is notified to the FDB unit 21 to perform address search, and based on the destination address search result obtained from the FDB unit 21, the port to be sent (ring connection port 22 or East ring port) 24) is selected. If the selected port is not blocked, the frame is transferred to the frame multiplexing control unit 13 together with the selected port information (selected port information).

In addition, the West port I / F unit 15 outputs the ERP control frame received from the in-ring failure notification transmission unit 194 to the PHY unit 17 in the transmission process. Further, when a frame is received from the frame multiplexing control unit 13 in a state where the West ring port 23 is not blocked, this frame is output to the PHY unit 17.

The East port I / F unit 16 transmits / receives a frame to / from the PHY unit 18. In the reception process, first, the validity of the received frame is confirmed. If it is not valid (there is an error), the frame is discarded. If it is valid, the type of the received frame is confirmed. In the case of an ERP control frame, arrival information indicating arrival of the ERP frame is output to the in-ring failure detection unit 191 and control information included in the ERP frame is included. Is output to the in-ring failure notification receiving unit 192. Also, information for address learning is extracted from the received frame and output to the FDB unit 21. Further, the destination of the received frame is notified to the FDB unit 21 to perform address search, and based on the search result of the destination address obtained from the FDB unit 21, the port to be sent (ring connection port 22 or West ring port) 23) is selected. If the selected port is not blocked, the frame is transferred to the frame multiplexing control unit 13 together with the selected port information (selected port information).

Further, the East port I / F unit 16 outputs the ERP control frame received from the in-ring failure notification transmission unit 194 to the PHY unit 11 in the transmission process. Further, when a frame is received from the frame multiplexing control unit 13 in a state where the East ring port 24 is not blocked, this frame is output to the PHY unit 18.

The PHY unit 17 extracts frame data from a received signal that is a communication medium signal arriving from a link connected on the West ring port 23 side, and transfers the frame data to the West port I / F unit 15 in the form of a frame. Further, the PHY unit 17 converts the frame data output from the West port I / F unit 15 into a communication medium signal, and transmits the communication medium signal from the West ring port 23.

The PHY unit 18 extracts frame data from a received signal that is a communication medium signal arriving from a link connected on the East ring port 24 side, and transfers the frame data to the East port I / F unit 16 in the form of a frame. The PHY unit 18 converts the frame data output from the East port I / F unit 16 into a communication medium signal, and transmits the communication medium signal from the East ring port 24.

The in-ring failure detection unit 191 of the SRP unit 19 detects a communication failure with an adjacent node in the ring A based on the reception result of the ERP control frame. For example, the reception status of the ERP control frame transmitted from each adjacent node is confirmed, and if the ERP control frame cannot be received for a certain period, a communication failure has occurred with the adjacent node that is the source of the ERP control frame. to decide. When a failure is detected, the failure detection is notified to the in-ring protection unit 193.

The in-ring failure notification receiving unit 192 confirms internal information (control information extracted from the ERP control frame) of the ERP control frame output from the West port I / F unit 15 and the East port I / F unit 16, and If failure information indicating the occurrence of a failure in A is included, the failure information is output to the in-ring protection unit 193. When the control information is in the ERP control frame transmitted from the redundant pair node (node a2), the received control information is output to the redundant pair node failure determination unit 203.

Whether the in-ring protection unit 193 needs to change the port block setting of the West ring port 23 and the East ring port 24 based on the output information from the in-ring failure detection unit 191 and the in-ring failure notification reception unit 192 to decide. When it is determined that the change is necessary, one or both of the west port I / F unit 15 and the east port I / F unit 16 are instructed to change the port blocking setting, and the blocking setting state is set to both routes. Notify the failure detection unit 204. Further, the in-ring protection unit 193 outputs information on a failure occurring in the ring A to the in-ring failure notification transmission unit 194.

The in-ring failure notification transmission unit 194 communicates the failure information in the ring notified from the in-ring protection unit 193 and the inter-ring route (inter-ring connection link 10-1) notified from the inter-ring route switching control unit 205. Based on the failure information, an ERP control frame is generated and sent to the West port I / F unit 15 and the East port I / F unit 16.

The MRP unit 20 that performs characteristic processing in the communication system of the present embodiment cooperates with the SRP unit 19 that performs path switching control in the ring to grasp the communication failure occurrence status in the ring and Based on this, it is determined whether or not the switching of the inter-ring route is necessary, and the inter-ring route is switched by changing the port blocking setting as necessary. In addition, when it is determined that the inter-ring path switching is necessary, it is determined that a failure requiring switching of the inter-ring path (corresponding to both path faults described later) has occurred. To notify. When there is a notification from the opposite ring connection node that a failure requiring switching of the inter-ring path has occurred, the port blocking setting is changed to switch the inter-ring path. That is, when the ring connection port 22 is closed at the time of notification, the block is released, and when the ring connection port 22 is not blocked, the block is set. Further, an FDB flush instruction for initializing (clearing) information stored in an FDB (Forwarding Data Base) and a buffer clear instruction for discarding frame data stored in the buffer are performed. Further, if necessary, the redundant pair node (node a2) is notified that a failure requiring switching of the inter-ring path has occurred.

Inter-ring failure notification transmission / reception unit 201 transmits / receives a multi-ring protection control frame to / from another connected ring network (adjacent ring). FIG. 3 is a flowchart illustrating an operation example of the inter-ring failure notification transmission / reception unit 201. The inter-ring failure notification transmission / reception unit 201 monitors whether failure information has been received from the adjacent ring (opposite ring connection node) via the ring connection port I / F unit 12 at a predetermined timing (step S11). When the information is received (step S11: Yes), the received failure information is transferred to the inter-ring path switching control unit 205 (step S12). The failure information is information transmitted in the multi-ring protection control frame, and a communication failure occurs between the two neighboring nodes in the ring at the failure information transmission source node (ring connection node). Indicates that they occur at the same time (referred to as a double-path failure). When failure information is not received (step S11: No), step S12 is skipped. Also, the inter-ring failure notification transmission / reception unit 201 is notified of the occurrence of both-path failures from the inter-ring route switching control unit 205 (receives a multi-ring protection control frame including failure information) (step S13: Yes). Then, a multi-ring protection control frame including failure information for notifying the occurrence of both path failures is generated and transferred to the ring connection port I / F unit 12 (steps S14 and S16). If the occurrence of both path failures is not notified (step S13: No), a multi-ring protection control frame for confirming the state of the inter-ring path is generated and transferred to the ring connection port I / F unit 12. (Steps S15 and S16).

Inter-ring failure detection unit 202 detects a communication failure with a ring connection node (opposite ring connection node) of an adjacent ring. FIG. 4 is a flowchart showing an operation example of the inter-ring failure detection unit 202. The inter-ring failure detection unit 202 confirms whether or not a multi-ring protection control frame has been received from the ring connection port I / F unit 12 (step S21), and if received (step S21: Yes), the inter-ring connection link 10- 1 is determined to be normal. If not received (step S21: No), the elapsed time since the previous reception is confirmed, and it is determined whether the non-reception time has timed out (step S22). When the elapsed time does not reach the predetermined value, that is, when it does not correspond to the non-reception time timeout (step S22: No), the process returns to the reception confirmation process of the multi-ring protection control frame. If the elapsed time has reached the predetermined value (step S22: Yes), it is determined that a communication failure has occurred with the opposite ring connection node, and the occurrence of an inter-ring communication failure is notified to the inter-ring path switching control unit 205. (Steps S23 and S24).

The redundant pair node failure determination unit 203 checks whether a communication failure has occurred between the redundant pair node (node a2) and its adjacent node (node a3 which is an adjacent node on the same ring). FIG. 5 is a flowchart illustrating an operation example of the redundant pair node failure determination unit 203. The redundant pair node failure determination unit 203 monitors whether an ERP control frame from the redundant pair node has been received via the in-ring failure notification reception unit 192 (step S31). When the ERP control frame is received from the redundant pair node, the intra-ring fault notification receiving unit 192 transfers the ERP control frame to the redundant pair node fault determining unit 203 and the in-ring protection unit 193. When the ERP control frame is received (step S31: Yes), the redundant pair node failure determination unit 203 confirms the included information, and between the redundant pair node and its adjacent node (adjacent node different from the own node). It is confirmed whether a communication failure has occurred (step S32). If no failure has occurred (step S32: No), the process is terminated. On the other hand, if a failure has occurred (step S32: Yes), the inter-ring path switching control unit 205 is notified that a failure has occurred in the redundant pair node (step S33). In addition, when the redundant pair node failure determination unit 203 does not receive an ERP control frame from the redundant pair node (step S31: No), the redundant pair node failure determination unit 203 checks the elapsed time since the previous ERP control frame was received and sets an ERP control frame non-reception timeout. It is determined whether or not this is the case (step S34). When the ERP control frame non-reception timeout is met, that is, when the ERP control frame is not received from the redundant pair node for a predetermined period (step S34: Yes), a failure occurs in the redundant pair node or the link with the redundant pair node. It is determined that it has occurred, and the fact is notified to the inter-ring path switching control unit 205 (steps S35 and S36). When the ERP control frame reception interval from the redundant pair node is within the specified range (step S34: No), reception monitoring of the ERP control frame from the redundant pair node is continued.

The both-path failure detection unit 204 checks whether or not a communication failure has occurred simultaneously with two adjacent nodes in its own ring (whether or not both-route failures have occurred). FIG. 6 is a flowchart illustrating an operation example of the both-path failure detection unit 204. Both path failure detection units 204 obtain the ring port blocking setting information from the in-ring protection unit 193 and confirm it. That is, it is confirmed whether or not both the West ring port 23 and the East ring port 24 are closed (step S41). If one or more ports are not blocked (step S41: No), it is determined that both path failures have not occurred, and the process ends. On the other hand, when both the West ring port 23 and the East ring port 24 are closed (step S41: Yes), it is determined that a failure has occurred in both paths, and the fact is notified to the inter-ring path switching control unit 205. (Steps S42 and S43).

The inter-ring path switching control unit 205 is based on information received from or notified from the inter-ring fault notification transmission / reception unit 201, the inter-ring fault detection unit 202, the redundant pair node fault determination unit 203, and the both-path fault detection unit 204. The switching control of the inter-ring route is performed. The inter-ring path switching control unit 205 issues a ring connection port blockage setting instruction to the ring connection port I / F unit 12, a buffer clear instruction to the buffer control unit 14, and a flush instruction to the FDB unit 21. In addition, when a both-path failure occurrence notification is received from the opposite ring connection node, both-path failure occurrence information indicating this is notified to the in-ring failure notification transmission unit 194. The FDB flush instruction for each node in the ring that is required when the path is switched is when the path in the ring is changed due to the switching of the path between rings (when the port blockage setting in the ring changes) Only)), an FDB flush instruction is issued to other nodes in the same ring. At this time, an FDB flush instruction may be performed using an ERP control frame.

FIG. 7 is a flowchart showing an operation example of the inter-ring path switching control unit 205. As shown in the figure, the inter-ring path switching control unit 205 determines whether a communication failure has occurred in the inter-ring path (step S51), whether both path faults have occurred in its own ring, or both path faults have occurred from adjacent rings. It is confirmed whether or not a notification has been received (step S52), whether or not a failure has occurred in the own ring or a node failure has been received (step S53), and a control operation is performed according to the confirmation result.

Specifically, first, it is confirmed whether or not a communication failure has occurred in the path between rings (step S51). When a communication failure is detected (step S51: Yes), a ring connection port blockage setting instruction is sent to the ring connection port I / O. Go to F12 to close the ring connection port 22 (step S58). In addition, the buffer controller 14 is instructed to clear the frame buffer, the FDB unit 21 is instructed to flush, and the intra-ring failure is notified to the redundant pair node that a communication failure has occurred in the inter-ring path. An instruction is given to the notification transmitter 194 (step S59).

If no communication failure is detected in the inter-ring route (step S51: No), then it is confirmed whether both route failures have occurred in the own ring or whether a notification of both route failures has been received from the adjacent ring. (Step S52). When both path failures occur in the own ring, or when notification of occurrence of both path failures is received from the adjacent ring (step S52: Yes), a ring connection port blockage setting instruction is sent to the ring connection port I / F 12. The ring connection port 22 is closed (step S56). Further, the buffer control unit 14 is instructed to clear the frame buffer, and the FDB unit 21 is instructed to perform FDB flush (step S57). If both path failures occur in the adjacent ring, in step S57, an instruction (inter-ring failure notification transmission instruction) is issued to the intra-ring failure notification transmission / reception unit 194 so as to notify the redundant pair node to that effect. .

If both path failures have not occurred in the own ring and the occurrence notification of both path failures has not been received from the adjacent ring (step S52: No), then the failure occurrence notification in the own ring or It is confirmed whether a node failure notification has been received (step S53). When a notification of the occurrence of a failure in the own ring or a notification of a node failure is received (step S53: Yes), the buffer control unit 14 is instructed to clear the frame buffer, and the FDB unit 21 is instructed to FDB flush. (Step S54). Further, the ring connection port blockage release instruction is issued to the ring connection port I / F 12 to release the blockage of the ring connection port 22 (step S55). On the other hand, when neither a failure occurrence nor a node failure notification is received in the own ring (step S53: No), the operation ends without executing steps S54 and S55.

FIG. 8 is a diagram showing the ring connection port blockage control operation performed by the inter-ring path switching control unit 205 in a table format. In the four columns from the left in FIG. 8, that is, “inter-ring failure”, “both-path failure detection or adjacent ring failure notification reception”, “redundant pair node failure”, and “failure notification reception from redundant pair node” The state in which the occurrence is indicated by “◯”, and the state in which no occurrence has occurred is indicated by “X”. In the “port blocking” at the right end, the operation for blocking the port is indicated by “◯”, and the operation for releasing the port blocking is indicated by “×”.

Therefore, the inter-ring path switching control unit 205 detects a communication failure between rings (inter-ring failure), detects both path failures, or receives a failure occurrence notification from an adjacent ring. 22 is closed. The failure notification from the adjacent ring is a notification of failure of both paths in the opposite ring connection node (that is, the node b1) connected by the inter-ring connection link 10-1. In addition, when a communication failure between the rings and a failure in both paths are not detected and a failure occurrence (both route failure occurrence) is not received from the adjacent ring, a failure notification is issued from the redundant pair node (that is, node a2). Is received, or when the failure of the redundant pair node is detected, the ring connection port 22 is released from being blocked. Note that the failure notification from the redundant pair node is a notification indicating the occurrence of a communication failure between the rings.

In addition, when the ring connection port is blocked when there is no failure, for example, one of the redundant pair nodes is set as the master and the other is set as the slave. For example, one of the ring connection nodes in the same ring communicates with another ring (adjacent ring). At this time, if a failure occurs on the slave side, the port on the slave side is set to block the port, and after the port on the master side releases the port block, the failure is recovered. (Return to the previous state) or maintain the previous setting state after recovery (do not return to the state before the failure occurred). It may be changed.

The FDB unit 21 determines the frame transfer destination based on the address learning information extracted from the received frame by the ring connection port I / F unit 12, the West port I / F unit 15, and the East port I / F unit 16, respectively. FDB (Forwarding Data Base), which is a database to be used at the time of creation, is created (including updates) and held. When the ring connection port I / F unit 12, the West port I / F unit 15 and the East port I / F unit 16 receive an address search request from the received frame, the FDB is searched to specify the transfer destination port. In addition, the requester of the specific result is notified. When an FDB flush instruction is received from the inter-ring path switching control unit 205, the FDB flush is executed to initialize the FDB.

Next, there are several failure occurrence patterns regarding the characteristic operation in the communication system of the present embodiment, that is, the port blocking control operation when a communication failure occurs in a ring network or a route connecting ring networks. An example will be described.

(1) Control operation when failure (both route failure) occurs in both paths in the ring of the ring connection node FIG. 9 is a diagram showing the control operation when both path failures occur. A case where a failure has occurred in a path on both sides of the node a1 of A (a link connecting the nodes a1 and a2, a link connecting the nodes a1 and a5) is shown. In the state immediately before the failure occurs, the ring connection port is blocked at the node a2 and the node b2, and each node of the ring A and each node of the ring B communicate with each other via the nodes a1 and b1. To do. The control operations S101 to S104 in this case will be described below.

<Operation of S101>
In the SRP unit 19 (see FIG. 2) of the node a1, the in-ring failure detection unit 191 monitors the reception status of the ERP control frame transmitted from each of the nodes a2 and a5 which are adjacent nodes in the own ring. If the ERP control frame cannot be received for a certain period of time, it is determined that a communication failure has occurred with the adjacent node that is the source of the ERP control frame. The detection result of the communication failure by the in-ring failure detection unit 191 is notified to the in-ring protection unit 193, and the in-ring protection unit 193 notified of the communication failure detection includes the West port I / F unit 15 and the East port I / F unit. 16 is instructed to block the ring port (West ring port 23 or East ring port 24) to the side where the communication failure is detected, and the port is blocked. At this time, the in-ring protection unit 193 notifies the both-path failure notification unit 204 of the MRP unit 20 of the information of the link where the communication failure is detected or the blocked ring port. When the in-ring failure detection unit 191 detects a communication failure between both the nodes a2 and a5, the West ring port 23 and the East ring port 24 are blocked by the control of the in-ring protection unit 193, and this is indicated by MRP. The both-path failure notification unit 204 of the unit 20 is notified, and as a result, the both-path failure notification unit 204 detects both-path failures. In this case, the MRP unit 20 notifies the inter-ring path switching control unit 205 that both path faults have been detected, and the inter-ring path switching control unit 205 notifies the inter-ring fault notification transmitting / receiving unit 201 that both path faults have occurred. Is notified to the ring connection node (opposite ring connection node) of the adjacent ring. Receiving this instruction, the inter-ring failure notification transmission / reception unit 201 generates a multi-ring protection control frame including information indicating the occurrence of both path failures, and transmits it to the node b1 of the adjacent ring. Further, the inter-ring path switching control unit 205 instructs the ring connection port I / F unit 12 to set the ring connection port 22 to be blocked, and blocks the port. In addition, an FDB flush instruction is issued to the FDB unit 21 to initialize the FDB, and an instruction is issued to the buffer control unit 14 to clear the frame buffer storing the received frame from the adjacent ring.

Note that FDB initialization is performed because the FDB that has been used until then cannot be used because the port blocking setting has been changed. If this is not performed, frame transfer will not be performed normally. The frame buffer is cleared to prevent the frame arrival order at the destination node from being reversed.

<Operation of S102>
When the node b1 receives from the node a1 a multi-ring protection control frame including information indicating the occurrence of both-path failures, a frame buffer for setting a ring connection port to be blocked and storing a received frame from an adjacent ring To clear. In addition, the redundant pair node indicates that communication via the inter-ring connection link 10-1 (see FIG. 1) connecting the node a1 is impossible, that is, communication with the adjacent ring is impossible. Notify node b2. Also, FDB flush is performed. As in the case where the node a1 performs the operation of S101, the ring connection port block setting, the frame buffer clearing, and the FDB flush processing are performed by the inter-ring path switching control unit 205 in the node b1. Accordingly, the ring connection port I / F unit 12, the buffer control unit 14, and the FDB unit 21 in the node b1 carry out. Notification to the node b2 which is the redundant pair node (notification that communication via the inter-ring connection link 10-1 is impossible) is performed according to the instruction from the inter-ring path switching control unit 205. This is performed by the internal failure notification transmitter 194. This notification is, for example, an ERP control frame including information indicating that communication via the inter-ring connection link 10-1 connecting the node a1 is impossible (communication with an adjacent ring is impossible). This is done by sending Notification may be made using other control frames.

<Operation of S103>
When the node b2 receives a notification from the redundant pair node b1 that communication with the adjacent ring is impossible, the node b2 clears the frame buffer storing the received frame from the adjacent ring, and the ring connection port. Release the blockage. Also, FDB flush is performed. The procedures of frame buffer clearing, ring connection port blocking control operation, and FDB flushing are the same as those of the nodes a1 and b1.

<Operation of S104>
For example, the node a2 confirms the reception status of the ERP control frame or the multi-ring protection control frame transmitted from the node a1 and the information stored in the ERP control frame. If it is determined that a failure has occurred in both paths, the frame buffer storing the received frame from the adjacent ring is cleared and the ring connection port is released from being blocked. Also, FDB flush is performed. The procedures of frame buffer clearing, ring connection port blocking control operation, and FDB flushing are the same as those of the nodes a1, b1, and b2.

Here, the operation of detecting the occurrence of both-path failure in the node a1 in which the node a2 is a redundant pair node will be described. In a state where no communication failure has occurred in the ring network, each node periodically transmits an ERP control frame for failure detection to each adjacent node, and cannot receive the ERP control frame transmitted from the adjacent node for a certain period of time. In this case, it is determined that a failure has occurred, and an ERP control frame for failure detection notification is transmitted to the adjacent node on the opposite side of the failure side. The node that has received the ERP control frame for failure detection notification transfers the frame if no failure has occurred in the link on the opposite side of the frame receiving this frame. The node that has blocked the port to avoid the loop releases the port blocking. Therefore, as described below, the node a2 can detect the occurrence of both-path failures in the node a1 that is the redundant pair node. For example, when a failure occurs between the nodes a1 and a5, the node a2 causes the failure between the nodes a1 and a5 when the ERP control frame for failure detection notification transmitted from the node a5 is received. recognize. After that, when the ERP control frame for detecting a failure from the node a1 cannot be received for a certain period of time, the occurrence of a failure between the node a1 and the own node is detected, and as a result, the occurrence of both-path failures at the node a1 is recognized. it can. On the other hand, when the failure occurrence between the node a1 and the own node is detected first, the own node and the node a5 transmit the ERP control frame for failure detection notification, and as a result, a port for loop avoidance Since the blockage is released, if a failure occurs between the nodes a1 and a5, a failure detection notification ERP control frame transmitted from the node a5 (a failure occurs between the nodes a1 and a5). The ERP control frame indicating that the two routes are faulty at the node a1 can be recognized. Note that the detection of the occurrence of both-path failures in the redundant pair node (node a1) is performed by the redundant pair node failure determination unit 203.

As described above, in the communication system of the present embodiment, even when a double path failure occurs in a ring connection node (ring connection node that does not block the ring connection port) in communication with another ring, Each ring connection node recognizes the occurrence of a failure, changes the ring connection port blocking setting, and switches the path between rings, so that communication between rings can be maintained.

(2) Control operation when a failure occurs in an inter-ring route FIG. 10 is a diagram showing a control operation when a failure occurs in an inter-ring route (a route connecting ring A and ring B). As an example, a case where a failure occurs in a link connecting the node a1 of the ring A and the node b1 of the ring B is shown. In the state immediately before the failure occurs, the ring connection port is blocked at the node a2 and the node b2, and each node of the ring A and each node of the ring B communicate with each other via the nodes a1 and b1. To do. The control operations S201 to S204 in this case will be described below.

<Operation of S201>
In the MRP unit 20 (see FIG. 2) of the node a1, the inter-ring failure detection unit 202 monitors the reception state of the multi-ring protection control frame transmitted from the node b1 which is the ring connection node of the adjacent ring, and is constant. If the multi-ring protection control frame cannot be received over time, it is determined that a communication failure has occurred with the node b1 that is the transmission source of the multi-ring protection control frame. The detection result of the communication failure by the inter-ring failure detection unit 202 is notified to the inter-ring route switching control unit 205, and the inter-ring route switching control unit 205 notified of the communication failure detection notifies the ring connection port I / F unit 12. Then, the ring connection port 22 is blocked by instructing the block connection setting of the ring connection port 22. In addition, an FDB flush instruction is issued to the FDB unit 21 to initialize the FDB, and an instruction is issued to the buffer control unit 14 to clear the frame buffer storing the received frame from the adjacent ring. Further, a failure occurs in the inter-ring route to the in-ring failure notification transmission unit 194 of the SRP unit 19 (the communication via the inter-ring connection link 10-1 becomes impossible). An instruction is given to notify a2). Upon receiving this instruction, the in-ring failure notification transmission unit 194 generates an ERP control frame including information indicating the occurrence of a failure in the inter-ring route, and transmits it to the node a2.

<Operation of S202>
When the node b1 detects a communication failure with the node a1 in the same procedure as the node a1, the node b1 performs FDB flush, frame buffer clear, and failure detection notification to the node b2.

<Operation of S203>
When the node a2 receives an ERP control frame including information indicating the occurrence of a failure in the inter-ring path from the node a1, the node a2 releases the blocking of the ring connection port. In addition, the frame buffer storing the received frame from the adjacent ring is cleared and FDB flush is performed.

<Operation of S204>
When the node b2 receives a failure occurrence notification on the inter-ring path from the node b1, the node b2 performs block connection release, frame buffer clear, and FDB flush in the same procedure as the node a2.

As described above, in the communication system according to this embodiment, even when a failure occurs in an inter-ring route (inter-ring connection link) that is being used for communication with another ring, this failure occurrence is connected to each ring. Since the node recognizes and changes the blocking setting of the ring connection port and switches the path between rings, communication between rings can be maintained.

(3) Control Operation When Ring Connection Node Fails FIG. 11 is a diagram showing a control operation when the ring connection node fails. As an example, FIG. 11 shows a case where the node a1 of ring A fails. In a state immediately before the failure of the node a1, the ring connection port is blocked at the node a2 and the node b2, and each node of the ring A and each node of the ring B communicate with each other via the nodes a1 and b1. And The control operations S301 to S303 in this case will be described below.

<Operation of S301>
When the node a1 fails, the node b1 detects a communication failure with the node a1 in the same procedure as the control operation S202 described above, and performs FDB flush, frame buffer clear, and failure detection notification to the node b2.

<Operation of S302>
When the node b2 receives a failure occurrence notification on the inter-ring path from the node b1, the node b2 performs block connection port block release, frame buffer clear, and FDB flush in the same procedure as in the control operation S204 described above.

<Operation of S303>
The node a2, which is a redundant pair node of the node a1, determines that a communication failure has occurred with the node a1 when the ERP control frame from the node a1 cannot be received for a certain period of time. It is not possible to determine whether or not the node a1 has failed. However, if the node a1 fails, a communication failure occurs between the nodes a1 and a5. Therefore, a communication failure between the failure detection node a1 is detected. After that, the ERP control frame for failure detection notification transmitted from the node a5 is received. Therefore, when the node a1 fails, the node a2 performs the same control as the above-described case where the both-path failure has occurred in the node a1, and from the ring connection port block release, FDB flush, and adjacent ring Clear the frame buffer that stores the received frames.

As described above, in the communication system according to the present embodiment, even when a ring connection node in communication with another ring fails and a communication failure occurs, the other ring connection node recognizes this failure. Since the ring connection port blockage setting is changed and the path between rings is switched, communication between rings can be maintained.

As described above, in the communication system according to the present embodiment, the ring connection node is in the same ring (own ring) in a state where the ring connection port is not blocked (communication with an adjacent ring). When a signal received from an adjacent ring cannot be transferred to another node due to a communication failure (when both path failures occur), the failure occurrence is notified to the ring connection node of the adjacent ring and the port block setting in the adjacent ring is changed. I decided to let them. In addition, when the occurrence of both path failures is notified from the adjacent ring, the ring connection port is blocked, and the failure occurrence is notified to other ring connection nodes (redundant pair nodes) in the same ring to block the ring connection port. It was decided to cancel. In addition, the communication status between rings is monitored by sending and receiving a control signal (multi-ring protection control frame) for detecting a failure with the ring connection node of the adjacent ring, and if a communication failure is detected, a failure occurs. Is notified to the other ring connection node (redundant pair node) in the same ring to release the block of the ring connection port. On the other hand, in a state where the ring connection port is blocked (not communicating with the adjacent ring), it is monitored whether both path failures have occurred in the redundant pair node, and both path failures have been detected. Decided to release the block on the ring connection port and start communication with the adjacent ring. Further, even when a redundant pair node is notified that a double path failure has occurred in the ring connection node of the adjacent ring, the ring connection port is released and communication with the adjacent ring is started. As a result, each ring connection node can recognize and change the port blocking setting when communication between rings becomes impossible. As a result, communication between rings can be continued even in the case where a failure has occurred that makes communication between rings impossible.

Embodiment 2. FIG.
In the first embodiment, the control operation when various failures occur is shown, but in this embodiment, the control operation when recovering from a failure is shown. The configuration of the communication system is the same as that of the first embodiment (see FIG. 1).

FIG. 12 is a diagram illustrating a configuration example of the ring connection node according to the second embodiment. As an example, a configuration example of the node a1 that is a ring connection node is illustrated. The configuration of other ring connection nodes is the same. The ring connection node of the present embodiment is obtained by replacing the MRP unit 20 of the ring connection node (see FIG. 2) described in the first embodiment with an MRP unit 20a, and further adding a path switching execution instruction unit 25 at the time of failure recovery. It is. Since the parts other than the MRP unit 20a and the failure recovery path switching execution instruction unit 25 are the same as the ring connection node described in the first embodiment, the description other than these will be omitted.

The MRP unit 20a includes an inter-ring failure notification transmission / reception unit 201a, an inter-ring failure detection unit 202a, a redundant pair node failure determination unit 203a, a both-path failure detection unit 204a, and an inter-ring route switching control unit 205a. The MRP unit 20 described in the first embodiment has a function of detecting the occurrence of a failure that makes inter-ring communication impossible and changing the blocking setting of the ring connection port. The MRP unit 20a further has a function of detecting recovery from a failure related to inter-ring communication and changing the blocking setting of the ring connection port.

The inter-ring failure notification transmission / reception unit 201a is obtained by adding the following functions to the inter-ring failure notification transmission / reception unit 201 described in the first embodiment. The added function is to generate a multi-ring protection control frame including recovery information indicating that recovery from both path failures is received when ring path recovery control notification is received from the inter-ring path switching control unit 205a. When a multi-ring protection control frame including a function for transmitting to the adjacent ring via the port I / F unit 12 and recovery information indicating recovery from both path failures is received from the adjacent ring, the recovery information is transmitted between the rings. This function is output to the route switching control unit 205a.

The inter-ring path failure detection unit 202a is obtained by adding the following functions to the inter-ring path failure detection unit 202 described in the first embodiment. The added function is to continue to receive multi-ring protection control frames from the adjacent ring even after detecting an inter-ring path failure (communication failure with the adjacent ring). When the protection control frame is received, this is a function for notifying the inter-ring path switching control unit 205a of recovery from the failure.

Even when the ring connection port is blocked, it is possible to set a part of the control signal such as the multi-ring protection control frame to pass through the blocked ring connection port. In this embodiment, it is assumed that at least a multi-ring protection control frame is passed through a blocked port.

The redundant pair node failure determination unit 203a is obtained by adding the following functions to the redundant pair node failure determination unit 203 described in the first embodiment. The added function is that when a failure between a redundant pair node and an adjacent node is detected and control information indicating recovery from this failure is received via the in-ring failure notification receiving unit 192, a failure on the redundant pair node side is detected. A function for notifying the inter-ring path switching control unit 205a that recovery has occurred, and when the control information indicating recovery from failure is received via the in-ring failure notification receiving unit 192, it indicates that the redundant pair node has recovered from failure. The function for notifying the inter-path switching control unit 205a, the redundancy when the control information indicating that the ring connection port blocking setting has been changed on the redundant pair node side is received via the intra-ring fault notification receiving unit 192 due to failure recovery. This is a function for notifying the inter-ring path switching control unit 205a that the ring connection port blocking setting has been changed on the pair node side.

The both-path failure detection unit 204a is obtained by adding the following functions to the both-path failure detection unit 204 described in the first embodiment. The added function is that the ERP unit 19 cancels the port block setting of either the West ring port 23 or the East ring port 24 in a state in which both path failures occur, and this is indicated by the in-ring protection unit 193. When notified, it is a function that determines that the failure has occurred from both path failures and notifies the inter-ring path switching control unit 205a to that effect.

The inter-ring path switching control unit 205a is obtained by adding the following functions to the inter-ring path switching control unit 205 described in the first embodiment. The added function is that when information is instructed to perform path switching at the time of failure recovery when the recovery from a failure is held in the path switching execution instruction unit 25 at the time of failure recovery, an inter-ring failure notification transmission / reception unit 201a, This is a function for performing switching control of the inter-ring path based on the information received from the information received from the inter-ring path fault detection unit 202a, the redundant pair node fault determination unit 203a, and both path fault detection units 204a or the notified contents.

FIG. 13 is a diagram showing the ring connection port blocking control operation performed by the inter-ring path switching control unit 205a in the form of a table. The two columns of “Recovery notification received from redundant pair node” and “Route switching execution notification received from redundant pair node” are added to FIG. 8 used in the description of the first embodiment.

The inter-ring path switching control unit 205a, when receiving the failure recovery information indicating the recovery notification from the failure from the redundant pair node, passes the received failure recovery information to the intra-ring failure notification transmission unit 194. In addition, route switching is performed by changing the ring connection port blocking setting based on the received failure recovery information, and route switching execution information indicating that the route switching has been performed based on the received failure recovery information. , It is passed to the failure notification transmitter 194 in the ring. The information passed to the in-ring failure notification transmission unit 194 is transmitted to the redundant pair node by using an ERP control frame.

The failure recovery path switching execution instruction unit 25 changes the blocking setting of the ring connection port 22 due to the occurrence of a failure and switches the inter-ring path, and then recovers from this failure and then the inter-ring path (ring connection port 22 blocking). Instruction information indicating whether to return the setting to the state before the failure occurs is stored, and an instruction according to this information is output to the inter-ring path switching control unit 205a. Whether or not to restore the setting when recovering from a failure may be set by the user from the outside.

Next, a characteristic operation in the communication system of the present embodiment, that is, a port blocking control operation when a communication failure occurring in a ring network or a route connecting ring networks is recovered is shown in FIG. Will be described.

FIG. 14 is a diagram showing a control operation when a failure occurring in the inter-ring path is recovered. As an example, a failure occurring in the inter-ring connection link 10-1 connecting the nodes a1 and b1 is shown. It shows the case of recovery. The control operations S401 to S404 in this case will be described below.

<Operation of S401>
The node a1 monitors whether or not a multi-ring protection control frame is transmitted from the node b1 even in a state where a communication failure (failure in an inter-ring path) occurs with the node b1, and the multi-ring When the protection control frame is received, it is determined that the failure has been recovered. Then, an ERP control frame including control information indicating that a failure in the inter-ring path has been recovered (referred to as failure recovery information) is transmitted to the node a2 that is the redundant pair node.

<Operation of S402>
Similarly to the node a1, the node b1 detects the failure recovery by receiving the multi-ring protection control frame from the node a1, and makes the ERP control frame including the failure recovery information indicating that the failure in the inter-ring path has been recovered redundant. It transmits to node a2 which is a pair node.

<Operation of S403>
When the node a2 receives the ERP control frame including the failure recovery information from the node a1, the node a2 clears the frame buffer storing the received frame from the adjacent ring and closes the ring connection port. Also, FDB flush is performed. Furthermore, an ERP control frame including control information (path switching execution information) indicating that the ring connection port has been blocked is transmitted to the redundant pair node node a1.

<Operation of S404>
Similarly to the node a2, the node b2, when receiving an ERP control frame including failure recovery information from the node b1, clears the frame buffer storing the received frame from the adjacent ring, blocks the ring connection port, and performs FDB flush. The ERP control frame including the path switching execution information is transmitted to the node b1 that is the redundant pair node.

In addition, when the node a1 receives the ERP control frame including the path switching execution information from the node a2 after transmitting the ERP control frame including the failure recovery information to the node a2 in the control operation S401 described above, Clear the frame buffer in which the received frame from the ring is stored, release the blocking of the ring connection port, and perform FDB flush.

Similarly, when the node b1 receives the ERP control frame including the path switching execution information from the node b2 after transmitting the ERP control frame including the failure recovery information to the node b2 in the control operation S402 described above, Clear the frame buffer in which the received frame from the ring is stored, release the blocking of the ring connection port, and perform FDB flush.

As described above, in the communication system according to the present embodiment, when the failure recovery is detected, the ring connection port blockage setting is returned to the state before the failure occurrence. Sometimes, the inter-ring connection link used for communication between the rings can be fixed. For this reason, for example, the bandwidth of an inter-ring connection link used in a normal state where no failure has occurred can be different from the bandwidth of an inter-ring connection link used in the event of a failure (the bandwidth of a link used in a normal state> failure). Link bandwidth to be used when it occurs).

Also, when performing path switching (ring connection port blocking setting change) at the time of failure recovery, the inter-ring connection port is always blocked in either redundant pair node, thus preventing the occurrence of loop frames. it can.

In this embodiment, only the operation at the time of failure recovery in the inter-ring path has been shown, but at the time of recovery from other failures as well, the operation is performed so that the inter-ring connection port is blocked in either redundant pair node. To do.

In the first and second embodiments, the case where ERP is used as the protection technique for the route in the ring network has been described. However, another protection technique may be applied. For example, when the RPR (Resilient Packet Ring) is used, the present invention is applied, and path switching when both paths in the ring fail and path switching between rings when a failure occurs can be performed. In failure notification between redundant pair nodes, failure information may be transmitted using an RPR control frame. Redundant pair node failure detection information can also be detected from RPR protection information.

As described above, the communication system according to the present invention is useful for a communication system including a plurality of ring networks, and is particularly suitable when a connection link between ring networks is a dual system.

a1, a2, a3, a4, a5, b1, b2, b3, b4, b5 Nodes 10-1, 10-2

Inter-ring link

11, 17, 18 PHY unit 12 Ring connection port I / F unit 13 Frame multiplexing control Unit 14 Buffer control unit 15 West port I / F unit 16 East port I / F unit 19 Single ring protection unit (SRP unit)
20, 20a Multi-ring protection part (MRP part)
21 FDB unit 22 Ring connection port 23 West ring port 24 East ring port 191 In-ring failure detection unit 192 In-ring failure notification reception unit 193 In-ring protection unit 194 In-ring failure

notification transmission unit

201, 201a Inter-ring failure notification transmission /

reception unit

202 , 202a Inter-ring

failure detection unit

203, 203a Redundant pair node failure determination unit 204, 204a Both-path

failure detection unit

205, 205a Inter-ring path switching control unit

Claims

A communication system that includes a plurality of ring networks composed of a plurality of nodes, each ring network having two adjacent nodes as ring connection nodes and physically connected to other ring networks via these ring connection nodes Because
Each ring network
Communicate with the adjacent ring network via one of the ring connection nodes,
When a communication failure occurs between a ring connection node in communication with an adjacent ring network and both of the two adjacent nodes in the ring network at the same time, the adjacent ring network is notified of the occurrence of both path failures indicating that fact. And switching ring connection nodes that communicate with adjacent ring networks,
A communication system characterized by switching a ring connection node that communicates with an adjacent ring network even when notification of both path failure occurrences is received from the adjacent ring network.
Each ring connection node
When communicating with an adjacent ring network and receiving notification of both path failures from the adjacent ring network, it is determined that communication with the adjacent ring network is no longer possible. Notify the node,
2. When not communicating with an adjacent ring network, communication with the adjacent ring network is started when notified that communication with the adjacent ring network is impossible. The communication system according to 1.
Each ring connection node
If you are not communicating with the adjacent ring network,
Based on the communication state with the other ring connection node in communication with the adjacent ring network and the communication failure detection result by the other node in the own ring network, the other ring connection node is both of the two adjacent nodes. 2. The communication system according to claim 1, wherein communication with the adjacent ring network is started when it is determined whether or not communication with the adjacent ring network is possible. .
A ring connection node communicating with an adjacent ring network
Monitors the communication status with the adjacent ring network, and if a communication failure occurs, terminates communication with the adjacent ring network and notifies the other ring connection nodes in the own ring network of the communication failure. ,
A ring connection node that is not communicating with an adjacent ring network
2. The communication system according to claim 1, wherein communication with an adjacent ring network is started when the notification of the occurrence of the communication failure is received from another ring connection node in the own ring network.
Each ring connection node
The frame buffer for storing received frames from the adjacent ring network is cleared when communication with the adjacent ring network is started and when communication with the adjacent ring network is ended. Item 12. The communication system according to Item 1.
Each ring connection node
A database for storing information on a forwarding destination of a received frame is initialized when communication with an adjacent ring network is started and when communication with an adjacent ring network is terminated. The communication system according to 1.
If the route used for communication between ring networks has been changed due to the occurrence of a communication failure, when recovery from the communication failure is detected, the route used for communication between ring networks is returned to the state before the occurrence of the communication failure. The communication system according to any one of claims 1 to 6, wherein the communication system is returned.
It includes a plurality of ring networks composed of a plurality of nodes, and each ring network is physically connected to another ring network through two adjacent nodes and uses one of the two nodes. A communication device that operates as a ring connection node that is a node physically connected to another ring network in a communication system that communicates with another ring network,
In-ring communication monitoring means for monitoring the communication state between each adjacent node in the own ring network;
An inter-ring communication state that is a communication state between adjacent ring networks is monitored, and adjacent ring networks are connected based on the monitoring result by the intra-ring communication monitoring means and the monitoring result of the inter-ring communication state. A port control means for performing blockage control of the existing port;
When the port control means closes the port, a notification means for notifying other ring connection nodes to that effect;
A communication apparatus comprising:
The communication apparatus according to claim 8, wherein the port control unit further performs port blockage control based on a notification content from a notification unit of another ring connection node.
When the monitoring result by the in-ring communication monitoring means indicates that a communication failure occurs simultaneously with each adjacent node, and the port is not blocked,
The notifying means notifies an adjacent ring network that the port is blocked;
The communication apparatus according to claim 8, wherein the port control unit closes the port after the notification by the notification unit is completed.
When a notification is received from the ring connection node of an adjacent ring network that the port connected to the ring network is blocked,
The port control means closes a port connecting adjacent ring networks,
The communication device according to claim 8, wherein the notification unit performs the notification to another ring connection node of the own ring network.
If you receive a notification from another ring connection node of your ring network that the port connecting the adjacent ring network is blocked,
The communication device according to claim 8, wherein the port control unit releases the blockage of the port.
If the port control means detects that a communication failure has occurred with an adjacent ring network, the port control means closes the port,
The communication device according to claim 8, wherein the notifying unit notifies other ring connection nodes of the ring network that the port is blocked.
When the port control means changes the port blocking setting,
The communication apparatus according to claim 8, wherein a frame buffer that stores a reception frame from an adjacent ring network is cleared.
When the port control means changes the port blocking setting,
The communication apparatus according to claim 8, wherein a database for storing information on a transfer destination of a received frame is initialized.
The port control means detects recovery from a communication failure when the port has already been blocked due to a communication failure with an adjacent node in the own ring network or a communication failure with an adjacent ring network. Or when notified of recovery from communication failure, release the port blockage,
The notifying unit notifies other ring connection nodes of the own ring network that the port block is released before the port control unit releases the port block. Item 16. The communication device according to any one of Items 8 to 15.
A communication system that includes a plurality of ring networks including a plurality of nodes, and each ring network is physically connected to other ring networks via these ring connection nodes, with two adjacent nodes as ring connection nodes A communication method in which
As processing executed by each ring connection node,
In-ring communication monitoring step for monitoring the communication state between each adjacent node in the own ring network;
An inter-ring communication monitoring step for monitoring a communication state between adjacent ring networks;
A determination step of determining whether or not it is necessary to change the blocking setting of a port connected to an adjacent ring network based on the monitoring result in the intra-ring communication monitoring step and the monitoring result in the inter-ring communication monitoring step When,
A setting change step of changing the blockage setting of the port when it is determined that a change is required in the determination step;
A notification step for notifying other ring connection nodes of the determination result when it is determined that the change is necessary in the determination step;
A communication method comprising:
When receiving a notification from the ring connection node of an adjacent ring network that the port connected to the ring network is blocked, the port connecting the adjacent ring network is blocked and the port is blocked. Port blocking step for notifying other ring connection nodes of the ring network,
The communication method according to claim 17, further comprising:
Port block release step for releasing the block of the port connected to the adjacent ring network when the other ring connection node is notified that the port connected to the adjacent ring network is blocked ,
The communication method according to claim 18, further comprising: