A method and device for forwarding a message in a resilient packet ring

Abstract

The application provides a resilient packet ring message forwarding method and device. The method comprises the following steps: receiving a cross-device link aggregation announcement message through a resilient packet ring network; allocating a RPR cross-device link aggregation identifier for a RPR MAC address of a member node carried by the cross-device link aggregation announcement message; receiving an Ethernet unicast message; when a destination MAC address of the received Ethernet unicast message corresponds to a RPR cross-device link aggregation identifier; performing hash calculation according to the Ethernet unicast message, taking a RPR MAC address of a member node corresponding to a result of the hash calculation as a destination RPR MAC address, taking a RPR MAC address of the current node as a source RPR MAC address, and encapsulating the received Ethernet unicast message into a RPR unicast message; and sending the RPR unicast message through an out port of a shortest RPR path reaching the destination RPR MAC address of the RPR unicast message.

Description

Technical Field

[0001] This application relates to communication technology, specifically a flexible packet ring message forwarding method and device. Background Technology

[0002] RPR (Resilient Packet Ring) is a new type of MAC (Media Access Control) protocol that uses RPR MAC layer frame encapsulation to achieve transparent Ethernet over RPR transmission.

[0003] Within a single RPR ring, the direction in which an RPR node sends RPR data packets clockwise is ring 0, also known as the Outer Ring; the direction in which an RPR node sends RPR packets counterclockwise is ring 1, also known as the Inner Ring. RPR nodes send RPR data packets on ring 0 and receive them on ring 1 via eastward physical ports; conversely, they receive RPR data packets on ring 0 and send them on ring 1 via westward physical ports.

[0004] The east-west physical ports and west-facing physical ports of each RPR node form the RPR logical port. The RPR chip sends RPR packets to the RPR logical port for lower-loop forwarding, and the switching chip sends Ethernet packets to the RPR logical port for upper-loop forwarding. Within the RPR node, RPR packets are sent by the RPR chip to the switching chip through the internal physical Ethernet port for lower-loop forwarding; Ethernet packets are sent by the switching chip to the RPR chip through the internal port for upper-loop forwarding.

[0005] However, RPR messages in the RPR network are all uploaded and unloaded by a single RPR node, which causes devices accessing the RPR network through a faulty node to be unable to carry their data messages normally through the RPR network.

[0006] DRNI (Distributed Resilient Network Interconnect) is a cross-device link aggregation technology based on the IEEE P802.1AX protocol. DRNI virtualizes two physical devices into one to achieve cross-device link aggregation, thereby providing device-level redundancy protection and traffic load balancing. Therefore, it is necessary to solve how to provide redundant forwarding processing through DRNI's cross-device link mining and RPR node loop-on or loop-off forwarding. Summary of the Invention

[0007] The purpose of this application is to provide a method and device for resilient packet ring forwarding (RPR) networks, which provides redundancy protection and load balancing for packet forwarding.

[0008] To achieve the above objectives, this application provides a method for forwarding Resilient Packet Ring (RPR) packets. The method includes: receiving cross-device link aggregation (CPR) announcement messages through a Resilient Packet Ring (RPR) network; assigning RPR CPR identifiers to the RPR MAC addresses of member nodes carried in the CPR announcement messages; receiving Ethernet unicast packets; when the downstream node identifier corresponding to the destination MAC address of the received Ethernet unicast packet is an RPR CPR identifier; performing a hash calculation on the received Ethernet unicast packet, using the RPR MAC address of the member node corresponding to the hash calculation result as the destination RPR MAC address and the RPR MAC address of the current node as the source RPR MAC address, encapsulating the received Ethernet unicast packet into a first RPR unicast packet; and sending the first RPR unicast packet through the egress port of the shortest RPR path leading to the destination RPR MAC address of the first RPR unicast packet.

[0009] To achieve the above objectives, this application also provides a method for forwarding Resilient Packet Ring (RPR) packets. The device includes: a Resilient Packet Ring (RPR) processing unit for receiving cross-device link aggregation (CPR) announcement messages through a RPR network; an aggregation module for assigning RPR CPR aggregation identifiers to the RPR MAC addresses of member nodes carried in the CPR announcement messages; a switching unit for receiving Ethernet unicast packets; when the destination MAC address of the received Ethernet unicast packet corresponds to the downstream node identifier of the RPR CPR aggregation identifier; the RPR processing unit performs a hash calculation based on the received Ethernet unicast packet, using the RPR MAC address of the member node corresponding to the hash calculation result as the destination RPR MAC address and the RPR MAC address of the current node as the source RPR MAC address, encapsulating the received Ethernet unicast packet into a first RPR unicast packet; and sending the first RPR unicast packet through the output port of the shortest RPR path leading to the destination RPR MAC address of the first RPR unicast packet.

[0010] The beneficial effect of this application is that by aggregating the RPR nodes of the RPR ring network into a virtual DRNI device, it can provide device-level redundancy protection and traffic load sharing, realize site protection of RPR devices, and realize service load sharing among RPR devices. Attached Figure Description

[0011] Figure 1 A flowchart illustrating an embodiment of the flexible packet ring message forwarding method provided in this application;

[0012] Figure 2A-2BThis is a schematic diagram of cross-device link aggregation notification provided in an embodiment of this application;

[0013] Figure 3 A schematic diagram illustrating an embodiment of RPR broadcast message forwarding provided in this application.

[0014] Figure 4 A schematic diagram illustrating an embodiment of RPR unicast message forwarding provided in this application.

[0015] Figure 5 This is a schematic diagram of an embodiment of the flexible packet ring message forwarding device provided in this application. Detailed Implementation

[0016] The following detailed description will be provided with reference to several examples illustrated in the accompanying figures. In this detailed description, numerous specific details are used to provide a comprehensive understanding of the present application. Known methods, steps, components, and circuits are not described in detail in the examples to avoid obscuring their meaning.

[0017] In the terminology used, the term "including" means including but not limited to; the term "containing" means including but not limited to; the terms "above," "within," and "below" include the number itself; the terms "greater than" and "less than" mean not including the number itself. The term "based on" means based on at least a portion of them.

[0018] Figure 1 A flowchart illustrating an embodiment of the flexible packet loop forwarding method provided in this application; the method includes:

[0019] Step 101: Receive cross-device link aggregation announcement messages through the Resilient Packet Ring (RPR) network;

[0020] Step 102: Assign an RPR cross-device link aggregation identifier to the RPR MAC address of the member node carried in the cross-device link aggregation announcement message;

[0021] Step 103: Receive Ethernet unicast messages;

[0022] Step 104: When the destination MAC address of the received Ethernet unicast message corresponds to the next link node identifier, which is the RPR cross-device link aggregation identifier;

[0023] Step 105: Perform a hash calculation on the received Ethernet unicast message, use the RPR MAC address of the member node corresponding to the hash calculation result as the destination RPR MAC address, and use the RPR MAC address of this node as the source RPR MAC address to encapsulate the received Ethernet unicast message into an RPR unicast message; send the RPR unicast message through the outgoing port of the shortest RPR path to the destination RPR MAC address of the RPR unicast message.

[0024] Figure 1 The beneficial effect of this embodiment is that by aggregating the RPR nodes of the RPR ring network into a virtual DRNI device, it can provide device-level redundancy protection and traffic load sharing, realize site protection of RPR devices, and realize service load sharing among RPR devices.

[0025] Figure 2A-2B This is a schematic diagram of cross-device link aggregation notification provided in an embodiment of this application. Figure 2A In RPR network 200, nodes A and B are virtualized into a single device via DRNI (Distributed Resilient Network Interconnect) to achieve cross-device link aggregation, providing device-level redundancy protection and traffic load balancing. Nodes A and B are distributed relay (DR) RPR nodes on RPR network 200.

[0026] The Ethernet ports of the switching chip at node A and the switching chip at node B are connected as IPP (Intra-Portal Port) via a physical Ethernet link as IPL (Intra-Portal Link). They forward DRCP packets and data packets and synchronize MAC address and ARP entries. Therefore, unlike IRF (Intelligent Resilient Framework) / stacked systems, they do not need to synchronize all the information of each member device. As a result, the coupling in the control plane is much smaller than that in stacking.

[0027] In addition to the IPL link, there is also a Keep-alive link between devices A and B. Figure 2A (Not shown), used to detect the status of the peer DR RPR node, that is, to perform dual-master detection when IPL link failure is performed by exchanging Keep-alive messages.

[0028] Port U1 of node A and port U2 of node B are DR (Distributed Relay interface) interfaces, both connected to network device 210 and belonging to the same DR group (Distributed-Relay group). Node A is the primary distributed aggregation RPR node device, and node B is the secondary distributed aggregation RPR node.

[0029] Figure 2AIn the RPR topology discovery process, when nodes A and B send ATD (Attribute Discovery) frames 201 and 202, they use the newly added TLV field in the ATD frame as an advertisement field carrying the DRNI member device address to carry the cross-device link aggregation of node A's RPR MAC A and node B's RPR MAC B.

[0030] Nodes C, D, E, and F receive ATD frames 201 and 202 respectively, allocating RPR Trunk identifiers to cross-device link aggregation nodes A and B. In this application, the RPR Trunk identifier is a local hardware resource of each RPR node. When each node receives the cross-device link aggregation announcement ATD frame, it starts allocating from the first RPR Trunk identifier to be allocated.

[0031] Figure 2A In the process, nodes C, D, E, and F allocate RPR Trunk 0 for cross-device link aggregation of nodes A and B based on the received ATD frames 201 and 202, and record the corresponding RPR MAC A and RPR MAC B for RPR Trunk 0.

[0032] Figure 2B In this configuration, nodes D and E are aggregated into a virtual device in the same manner as nodes A and B. Node E is the primary distributed aggregation RPR node, and node D is the backup distributed aggregation RPR node. During the RPR topology discovery process, when nodes E and D send ATD frames 203 and 204, they use the newly added TLV field in the ATD frame as an advertisement field carrying the DRNI member device address, to carry the cross-device link aggregation of node E's RPR MAC address E and node D's RPR MAC address D.

[0033] Based on the received ATD frames 203 and 204, nodes A and B allocate RPR Trunk 0 for cross-device link aggregation of nodes D and E and record the corresponding RPR MAC D and RPR MAC E for RPR Trunk 0.

[0034] Based on the received ATD frames 203 and 204, nodes C and F allocate RPR Trunk 1 for cross-device link aggregation of nodes D and E, and record the corresponding RPR MAC D and RPR MAC E for RPR Trunk 1.

[0035] Figure 3 A schematic diagram provided for the RPR broadcast message forwarding embodiment provided in this application;

[0036] Before terminal T1 sends a unicast data packet to terminal T2, if it does not find the MAC address entry corresponding to terminal T2's IP address in its table, it sends an ARP request packet 301.

[0037] When network device 220 receives ARP request message 301, it selects the member port of node E in the aggregation port of the link aggregation group connecting the DRNI system and sends ARP request message 301 to node E.

[0038] The switching chip of node E learns the MAC address table entry based on the source MAC address of ARP request message 301, and forwards ARP request message 301 to the RPR logical port for loop forwarding.

[0039] The RPR chip of node E determines that it is the primary distributed aggregation RPR node device and broadcasts an RPR broadcast message on RPR ring 200. The RPR chip of node E receives an ARP request message 301 through its internal port, encapsulates it into an RPR broadcast message 302, in which the source RPR MAC is the local node's RPR MAC address and the destination RPR MAC is the broadcast RPR MAC address, and sends them through the RPR ports connecting nodes F and D respectively.

[0040] When node D's RPR chip receives RPR broadcast message 302, it determines that the device's role is a standby distributed aggregation RPR node and that the source RPR MAC is the peer DR node. Therefore, it discards RPR broadcast message 302, meaning it does not perform the loop-down forwarding of RPR broadcast message 302, nor does it perform the loop-over forwarding of RPR broadcast message 302 on the RPR ring.

[0041] Node F's RPR chip receives RPR broadcast message 302 and forwards it to Node A through another RPR port on the RPR ring. A copy of RPR broadcast message 302 is then sent to the switching chip for ring-down processing. Node F's switching chip receives RPR broadcast message 302, looks up the corresponding RPR Trunk identifier RPR Trunk1 based on the RPR source MAC address, learns the MAC address table entry based on RPR Trunk 1 and the source MAC address of the inner ARP request message 301, removes the RPR header, and broadcasts it within VLAN 1000 (not shown in the diagram).

[0042] When node A's RPR chip receives RPR broadcast message 302, it determines that the device is the primary distributed aggregation RPR node and that the source RPR MAC is not the peer DR node. Then, it performs the loop-down forwarding of RPR broadcast message 302 and the loop-over forwarding of RPR broadcast message 302 on the RPR ring.

[0043] Node A sends an RPR broadcast message 302 through another RPR port on the RPR ring, and copies the RPR broadcast message 302 to the switching chip for ring-down processing. Upon receiving the RPR broadcast message 302, the switching chip of Node A looks up the corresponding RPR Trunk0 based on the RPR source MAC address. It then learns the MAC address table entry based on RPR Trunk0 and the source MAC address of the inner ARP request message 301, removes the RPR header, broadcasts it within VLAN 1000, and forwards the ARP request message 301 to network device 210. Finally, network device 210 forwards the ARP request message 301 to terminal T2.

[0044] When node B's RPR chip receives RPR broadcast message 302, it determines that its role is a standby distributed aggregation RPR node and that the source RPR MAC is not the peer DR node. Therefore, it does not perform the loop-down forwarding of RPR broadcast message 302, but instead performs the loop-over forwarding of RPR broadcast message 302.

[0045] Node B sends an RPR broadcast message 302 to Node C through another RPR port on RPR network 200.

[0046] Node C's RPR chip receives RPR broadcast message 302 and forwards it to Node D through another RPR port on RPR network 200. A copy of RPR broadcast message 302 is then sent to the switching chip for loopback processing. Node C's switching chip receives RPR broadcast message 302, looks up the corresponding RPR Trunk1 based on the RPR source MAC address, learns the MAC address table entry based on RPR Trunk1 and the source MAC address of the inner ARP request message 301, removes the RPR header, and broadcasts it within VLAN 1000 (not shown in the diagram).

[0047] Node D receives RPR broadcast message 302 from node C. If node D's RPR chip determines that its role is a standby distributed aggregate RPR node and that the source RPR MAC is the peer DR node, then it discards RPR broadcast message 302.

[0048] Figure 3 When terminal T2 sends an Ethernet broadcast message 321 in VLAN 1000, network device 210 receives the Ethernet broadcast message 321, selects a member port of node B in the aggregation port of the link aggregation group connecting to the DRNI system, and sends the Ethernet broadcast message 321 to node B.

[0049] The switching chip of node B learns the MAC address table entry based on the source MAC address of the Ethernet broadcast message 321, and forwards the Ethernet broadcast message 321 to the RPR logical port for loop forwarding; the internal port of the RPR chip of node B receives the Ethernet broadcast message 321, determines that the device's role is a backup distributed aggregation RPR node, encapsulates the Ethernet broadcast message 321 into an RPR broadcast message 322, in which the source RPR MAC is the RPR MAC address of this node, the destination RPR MAC is the broadcast RPR MAC, and sends it to the primary distributed aggregation RPR node A.

[0050] When Node A's RPR chip receives RPR broadcast message 322, it determines that its role is the primary distributed aggregation RPR node and that the source RPR MAC address is the peer DR node. Therefore, it performs loop forwarding of RPR broadcast message 322. Since the RPR MAC address is the peer DR node B, if Node A performs loop forwarding again, network device 210 will repeatedly receive Ethernet broadcast message 321, leading to a broadcast storm.

[0051] The RPR chip of node A sends an RPR broadcast message 322 through another RPR port of RPR network 200, thereby broadcasting the RPR broadcast message 322 on RPR network 200.

[0052] Node F's RPR chip receives RPR broadcast message 322 and forwards it to Node E through another RPR port on RPR network 200. A copy of RPR broadcast message 322 is then sent to the switching chip for loopback processing. Node F's switching chip receives RPR broadcast message 322, looks up the corresponding RPR Trunk0 based on the RPR source MAC address RPR MAC B, learns the MAC address entry based on RPR Trunk 0 and the source MAC address of the inner Ethernet broadcast message, removes the RPR header, and broadcasts it within VLAN 1000 (not shown in the diagram).

[0053] When node E's RPR chip receives RPR broadcast message 322, it determines that the device is the primary distributed aggregation RPR node and that the source RPR MAC is not the peer DR node. Then, it performs the loop-down forwarding of RPR broadcast message 302 and the loop-over forwarding of RPR broadcast message 322 on RPR network 200.

[0054] Node E sends an RPR broadcast message 322 through another RPR port on RPR network 200, and copies the RPR broadcast message 322 to the switching chip for loopback processing. Upon receiving the RPR broadcast message 322, the switching chip of Node E looks up the corresponding RPR Trunk identifier RPR Trunk0 based on the RPR source MAC address. It then learns the MAC address table entry based on RPR Trunk 0 and the source MAC address of the inner Ethernet broadcast message 321, removes the RPR header, broadcasts it within VLAN 1000, and sends the Ethernet broadcast message 321 to network device 220. Finally, network device 220 forwards the Ethernet broadcast message 321 to terminal T2.

[0055] When node D's RPR chip receives RPR broadcast message 322, it determines that its role is a backup distributed aggregation RPR node and that the source RPR MAC is not the peer DR node. Instead of performing loop-down forwarding of RPR broadcast message 302, it performs loop-over forwarding of RPR broadcast message 322 on RPR network 200.

[0056] Node D sends an RPR broadcast message 322 to Node C through another RPR port on RPR network 200.

[0057] Node C's RPR chip receives RPR broadcast message 322 and forwards it to Node B through another RPR port on RPR network 200. A copy of RPR broadcast message 322 is then sent to the switching chip for loopback processing. Node C's switching chip receives RPR broadcast message 322, looks up the corresponding RPR Trunk identifier RPR Trunk0 based on the RPR source MAC address, learns the MAC address table entry based on RPR Trunk0 and the source MAC address of the inner Ethernet broadcast message 321, removes the RPR header, and broadcasts it within VLAN 1000 (not shown in the diagram).

[0058] Node B receives RPR broadcast message 322 from Node C. Node B's RPR chip determines that its role is a standby distributed aggregate RPR node and that the source RPR MAC is its own RPR MAC address. Therefore, it discards RPR broadcast message 322.

[0059] Figure 4 A schematic diagram provided for the RPR unicast message forwarding embodiment provided in this application;

[0060] Node A and Node B synchronize MAC address and ARP entries via IPL; Node E and Node D synchronize MAC address and ARP entries via IPL.

[0061] Terminal T2 sends an Ethernet unicast message 411 to terminal T1. Network device 210 determines, based on the destination MAC address (MAC T1) of the Ethernet unicast message 411, that the outgoing interface is a link aggregation group connected to the DR group linking nodes A and B. Network device 210 performs a hash calculation for load balancing forwarding based on the Ethernet unicast message 411, and forwards the Ethernet unicast message 411 to node B according to the member link of the corresponding link aggregation group based on the hash value.

[0062] When Node B's switching chip receives Ethernet unicast message 411, it determines the destination MAC address MAC T1's output port as RPR Trunk0 based on the learned MAC address table entries and sends it to the RPR chip. Node B's RPR chip performs a hash calculation based on the message parameters of Ethernet unicast message 411, and selects the next node in RPR network 200 based on the RPR MAC address corresponding to the hash calculation result. For example, Node B determines that the hash value of the hash calculation result corresponds to RPR MAC D.

[0063] The RPR chip of node B encapsulates the Ethernet unicast message 411 into an RPR unicast message 412; the destination RPR MAC is RPR MAC D and the source RPR MAC is RPR MAC B; node B selects the RPR port on the shortest path to node D according to the topology of RPR network 200, and sends the RPR unicast message 412 to RPR node C on the shortest RPR path.

[0064] When the RPR chip of RPR node C receives RPR unicast message 412, it sends RPR unicast message 412 to RPR node D according to the RPR port on the shortest path of RPR node D.

[0065] When the RPR chip at RPR node D receives RPR unicast message 412, it determines that the destination RPR MAC address is the RPR MAC address of this device and forwards it to the switching chip. The switching chip at RPR node D removes the RPR encapsulation from RPR unicast message 412, looks up the corresponding RPR Trunk identifier RPR Trunk0 based on the RPR source MAC address RPR MAC B, refreshes the aging time of the learned MAC address table entry based on RPR Trunk 0 and the source MAC address T2 address of the inner Ethernet broadcast message, removes the RPR header, and finds that the outgoing port of the destination MAC address T2 in VLAN 1000 is the DR group.

[0066] RPR node D sends an Ethernet unicast message 411 to network device 220 through the local member port of the DR group. Network device 220 then sends the Ethernet unicast message 411 to terminal T2 based on the learned MAC address.

[0067] Figure 5 This is a schematic diagram of an embodiment of the resilient packet ring forwarding device provided in this application. The device 50 includes a switching unit, a resilient packet ring processing unit, a CPU, and a memory. The switching unit can be implemented using a switching chip, and the resilient packet ring processing unit can be implemented using an FPGA chip. The processor executes the aggregation module by running processor-executable instructions in the memory.

[0068] The resilient packet ring processing unit is used to receive cross-device link aggregation announcement messages through the resilient packet ring RPR network; the aggregation module is used to assign RPR cross-device link aggregation identifiers to the RPR MAC addresses of member nodes carried in the cross-device link aggregation announcement messages; the switching unit is used to receive Ethernet unicast messages; when the downstream node identifier corresponding to the destination MAC address of the received Ethernet unicast message is an RPR cross-device link aggregation identifier; the resilient packet ring processing unit performs a hash calculation based on the received Ethernet unicast message, uses the RPR MAC address of the member node corresponding to the hash calculation result as the destination RPR MAC address, and uses the RPR MAC address of the current node as the source RPR MAC address, encapsulates the received Ethernet unicast message into a first RPR unicast message; and sends the first RPR unicast message through the output port of the shortest RPR path to the destination RPR MAC address of the first RPR unicast message.

[0069] The resilient packet ring processing unit receives the second RPR unicast message through the RPR network. When the upstream node identifier corresponding to the source MAC address of the received second RPR unicast message is the RPR cross-device link aggregation identifier, the switching unit searches for a MAC address table entry based on the source MAC address of the received second RPR unicast message. If a match is found, the aging time of the matching MAC address table entry for the source MAC address of the received second RPR unicast message is refreshed. If no match is found, a MAC address table entry is learned based on the source MAC address of the received second RPR unicast message and the RPR cross-device link aggregation identifier.

[0070] The switching unit is used to receive Ethernet broadcast messages; the resilient packet ring processing unit is used to encapsulate the received Ethernet broadcast messages into a first RPR broadcast message and send it to the primary distributed aggregation RPR node when the device is a backup distributed aggregation RPR node; and to encapsulate the received Ethernet broadcast messages into a second RPR broadcast message and broadcast it on the RPR network when the device is a primary distributed aggregation RPR node.

[0071] The switching unit is used to receive the third RPR broadcast message through the RPR network; the flexible packet ring processing unit is used to discard the third RPR broadcast message when the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is a standby DR RPR node; and when the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is a primary DR RPR node, broadcast the third RPR broadcast message to adjacent RPR nodes through the RPR network.

[0072] The resilient packet ring processing unit is used to receive the fourth RPR broadcast message through the RPR network; when the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is a standby DR RPR node, it broadcasts the fourth RPR broadcast message to adjacent RPR nodes through the RPR network; when the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is the primary DR RPR node, it broadcasts the fourth RPR broadcast message to adjacent RPR nodes through the RPR network. The switching unit is used to convert the fourth RPR broadcast message into an Ethernet broadcast message; and broadcast it within the virtual local area network to which the converted Ethernet broadcast message belongs.

[0073] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for forwarding elastic packet loop messages, characterized in that, The method includes: Receive cross-device link aggregation announcement messages through the Resilient Packet Ring (RPR) network; Assign an RPR cross-device link aggregation identifier to the RPR MAC address of the member node carried in the cross-device link aggregation announcement message; Receive Ethernet unicast messages; When the destination MAC address of the received Ethernet unicast message corresponds to the next link node identifier, it is the RPR cross-device link aggregation identifier; The received Ethernet unicast message is hashed, and the RPR MAC address of the member node corresponding to the hash calculation result is used as the destination RPR MAC address, and the RPR MAC address of this node is used as the source RPR MAC address. The received Ethernet unicast message is then encapsulated into a first RPR unicast message. The first RPR unicast message is then sent through the egress port of the shortest RPR path to the destination RPR MAC address of the first RPR unicast message. Receive the second RPR unicast message through the RPR network; When the upstream node identifier corresponding to the source MAC address of the received second RPR unicast message is the RPR cross-device link aggregation identifier; The MAC address table entry is looked up based on the source MAC address of the received second RPR unicast message; When found, refresh the aging time of the matching MAC address entry in the source MAC address table of the received second RPR unicast message; If not found, the MAC address entry is learned based on the source MAC address of the received second RPR unicast message and the RPR cross-device link aggregation identifier.

2. The method according to claim 1, characterized in that, The method further includes: Receive Ethernet broadcast messages; When this device is a standby distributed aggregated RPR node, it encapsulates the received Ethernet broadcast message into a first RPR broadcast message and sends it to the primary distributed aggregated RPR node. When this device acts as a primary distributed aggregation RPR node, it encapsulates the received Ethernet broadcast message into a second RPR broadcast message and broadcasts it on the RPR network.

3. The method according to claim 1, characterized in that, The method further includes: Receive the third RPR broadcast message through the RPR network; If the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is a standby DR RPR node, the third RPR broadcast message is discarded. When the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is the primary DR RPR node, the third RPR broadcast message is broadcast to the adjacent RPR nodes through the RPR network.

4. The method according to claim 1, characterized in that, The method further includes: Receive the fourth RPR broadcast message through the RPR network; When the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is a standby DR RPR node, the fourth RPR broadcast message is broadcast to the adjacent RPR node through the RPR ring. When the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is the primary DR RPR node, the fourth RPR broadcast message is broadcast to the adjacent RPR nodes through the RPR network, and the fourth RPR broadcast message is converted into an Ethernet broadcast message; then it is broadcast within the virtual LAN to which the converted Ethernet broadcast message belongs.

5. A flexible packet ring message forwarding device, characterized in that, The device includes: The Resilient Packet Ring (RPR) processing unit is used to receive cross-device link aggregation announcement messages through the RPR network. The aggregation module is used to assign an RPR cross-device link aggregation identifier to the RPR MAC address of the member node carried in the cross-device link aggregation announcement message; A switching unit is used to receive Ethernet unicast packets; when the destination MAC address of the received Ethernet unicast packet corresponds to the downstream node identifier, the RPR cross-device link aggregation identifier is used. The resilient packet ring processing unit performs a hash calculation on the received Ethernet unicast packet, uses the RPR MAC address of the member node corresponding to the hash calculation result as the destination RPR MAC address, and uses the RPR MAC address of the current node as the source RPR MAC address to encapsulate the received Ethernet unicast packet into a first RPR unicast packet; and sends the first RPR unicast packet through the egress port of the shortest RPR path to the destination RPR MAC address of the first RPR unicast packet. The elastic packet ring processing unit receives a second RPR unicast message through the RPR network; when the upstream node identifier corresponding to the source MAC address of the received second RPR unicast message is the RPR cross-device link aggregation identifier; The switching unit searches for a MAC address entry in the table based on the source MAC address of the received second RPR unicast message. If a match is found, the aging time of the matching MAC address entry in the table is refreshed. If no match is found, the unit learns a MAC address entry in the table based on the source MAC address of the received second RPR unicast message and the RPR cross-device link aggregation identifier.

6. The device according to claim 5, characterized in that, The switching unit is used to receive Ethernet broadcast messages; The resilient packet ring processing unit is configured to, when the device is a standby distributed aggregate RPR node, encapsulate the received Ethernet broadcast message into a first RPR broadcast message and send it to the primary distributed aggregate RPR node; and when the device is a primary distributed aggregate RPR node, encapsulate the received Ethernet broadcast message into a second RPR broadcast message and broadcast it on the RPR network.

7. The device according to claim 5, characterized in that, The switching unit is used to receive third RPR broadcast messages through the RPR network; The resilient packet ring processing unit is configured to: discard the third RPR broadcast message when the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is a standby DR RPR node; and broadcast the third RPR broadcast message to adjacent RPR nodes through the RPR network when the upstream node of the third RPR broadcast message is the peer DR node of this node and this node is a primary DR RPR node.

8. The device according to claim 5, characterized in that, The elastic packet ring processing unit is used to receive the fourth RPR broadcast message through the RPR network; when the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is a backup DR RPR node, the fourth RPR broadcast message is broadcast to the adjacent RPR node through the RPR ring. When the upstream node of the fourth RPR broadcast message is not the peer DR node of this node and this node is the primary DR RPR node, the fourth RPR broadcast message is broadcast to the adjacent RPR nodes through the RPR network. The switching unit is used to convert the fourth RPR broadcast message into an Ethernet broadcast message and broadcast it within the virtual local area network to which the converted Ethernet broadcast message belongs.

Images