CN115277543B - A message forwarding method and device - Google Patents

本申请提供了一种报文转发方法及设备，其中该方法包括：设置弹性分组环分布式聚合组用以连接RPR环网的非分布式聚合RPR节点；设置分布式聚合组用以连接RPR环网的外部网络设备；设置广播报文过滤表项以禁止通过弹性分组环分布式聚合组的本地成员端口以及分布式聚合组的本地分布式聚合口转发通过内部控制链路IPP端口收到的以太网广播报文；当通过分布式聚合组接收以太网广播报文时，基于接收的以太网广播报文的源MAC地址以及分布式聚合组学习第一MAC地址表项；通过内部控制链路IPP端口发送接收的以太网广播报文，在接收的以太网广播报文所属虚拟局域网内广播；将接收的以太网广播报文封装为RPR广播报文在RPR环网转发。

The present application provides a message forwarding method and device, wherein the method comprises: setting a resilient packet ring distributed aggregation group to connect a non-distributed aggregation RPR node of an RPR ring network; setting a distributed aggregation group to connect an external network device of the RPR ring network; setting a broadcast message filtering table entry to prohibit forwarding an Ethernet broadcast message received through an internal control link IPP port through a local member port of the resilient packet ring distributed aggregation group and a local distributed aggregation port of the distributed aggregation group; when an Ethernet broadcast message is received through the distributed aggregation group, learning a first MAC address table entry based on a source MAC address of the received Ethernet broadcast message and the distributed aggregation group; sending the received Ethernet broadcast message through the internal control link IPP port, and broadcasting in a virtual local area network to which the received Ethernet broadcast message belongs; encapsulating the received Ethernet broadcast message into an RPR broadcast message and forwarding it in the RPR ring network.

Message forwarding method and device

Technical Field

The present application relates to communication technologies, and in particular, to an elastic packet ring network technology, and in particular, to a method and an apparatus for forwarding a message.

Background

RPR (RESILIENT PACKET RING ) is a MAC (MEDIA ACCESS Control) protocol that can run over SONET (Synchronous Optical Network )/SDH (Synchronous DIGITAL HIERARCHY, synchronous digital hierarchy), DWDM (DENSE WAVELENGTH Division Multiplexing ) and ethernet. The RPR uses RPR MAC layer frame encapsulation to implement Ethernet Over RPR transparent transmission. The ring structure and topology protection mechanism of the RPR are transparent to the forwarding process of the carried traffic and the access device.

When a certain link or a certain node on the RPR network fails, two nodes close to the failure point update their own topology databases firstly, then quickly send TP (Topology Protection ) frames to other nodes, the other nodes update the topology databases, and each node sends RPR messages according to the new topology to realize quick protection switching. However, the RPR messages of the RPR network are all a single RPR node, and once the paths of the RPR node serving as the aggregation device and two neighbor nodes on the RPR are failed, the messages which need to pass through the RPR node serving as the aggregation device on the RPR network cannot be forwarded.

DRNI (Distributed Resilient Network Interconnect ) is a cross-device link aggregation technology based on the IEEE P802.1AX protocol. The DRNI virtualizes two physical devices into one device to implement cross-device link aggregation, thereby providing device-level redundancy protection and traffic load sharing. The DRNI is mainly applied to the dual-homing access networking, and improves the reliability from a link level to a device level.

Disclosure of Invention

The application aims to provide message forwarding, and provides redundancy protection and load sharing for message forwarding of an RPR network through an aggregated elastic packet ring node.

The application provides a message forwarding method, which comprises the steps of setting a resilient packet ring distributed aggregation group to be connected with a non-distributed aggregation RPR node of a resilient packet RPR ring network, setting a distributed aggregation group to be connected with external network equipment of the RPR ring network, setting a broadcast message filtering table item to prohibit forwarding of an Ethernet broadcast message received through a local member port of the resilient packet ring distributed aggregation group and a local distributed aggregation port of the distributed aggregation group, receiving the Ethernet broadcast message through the distributed aggregation group, learning a first MAC address table item based on a source MAC address of the received Ethernet broadcast message and the distributed aggregation group, sending the received Ethernet broadcast message through an internal control link IPP port, broadcasting in a virtual local area network to which the received Ethernet broadcast message belongs, and packaging the received Ethernet broadcast message into the RPR broadcast message in the RPR forwarding ring network.

The application also provides a message forwarding device, which comprises a non-distributed aggregation RPR node for setting an elastic packet ring distributed aggregation group to connect with an elastic packet RPR ring network, an external network device for setting a distributed aggregation group to connect with the RPR ring network, a broadcast message filtering table item for prohibiting forwarding of an Ethernet broadcast message received through a local member port of the elastic packet ring distributed aggregation group and a local distributed aggregation port of the distributed aggregation group by an internal control link IPP port of a distributed aggregation RPR node connected with an opposite end, an Ethernet broadcast message received through the distributed aggregation group, a first MAC address table item based on a source MAC address of the received Ethernet broadcast message and the distributed aggregation group, a received Ethernet broadcast message sent through an internal control link IPP port and broadcasted in a virtual local area network to which the received Ethernet broadcast message belongs, and an RPR broadcast message encapsulated as the received Ethernet broadcast message is forwarded on the RPR ring network.

The application has the advantages that by aggregating the RPR nodes of the RPR ring network into virtual one DRNI virtual device, the device-level redundancy protection and the flow load sharing can be provided, the site protection of the RPR device can be realized, and the service load sharing among the RPR devices can be realized.

Drawings

Fig. 1 is a flowchart of an embodiment of a message forwarding method provided by the present application;

FIG. 2 is a schematic diagram of an RPR network with a distributed elastic network interconnection architecture according to the present application;

Fig. 3 is a schematic broadcast forwarding diagram of the ring network shown in fig. 2;

fig. 4 is a unicast forwarding schematic diagram of the ring network shown in fig. 2;

fig. 5 is a schematic diagram of an embodiment of a packet forwarding device according to the present application.

Detailed Description

A plurality of examples shown in the drawings will be described in detail. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the application. Well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the examples.

The terms "comprising" and "including" are used in the sense of including but not limited to, including the numbers "above", "within" and "below" and not including the numbers. The term "based on" means based at least in part on a portion thereof.

Fig. 1 is a flowchart of an embodiment of a message forwarding method provided by the present application, where the method includes the following steps:

step 101, setting RPR DR group to connect the non-distributed aggregation RPR node of elastic group RPR ring network;

102, setting a DR group to be connected with external network equipment of an RPR ring network;

step 103, setting a broadcast message filtering table item to prohibit forwarding of the Ethernet broadcast message received through the IPP port through the local member port of the RPR DR group and the local distributed aggregation port of the DR group;

Step 104, receiving an Ethernet broadcast message through a DR;

step 105, learning MAC address table items based on source MAC addresses and DR groups of received Ethernet broadcast messages;

Step 106, the received Ethernet broadcast message is sent through the IPP port, and broadcast is carried out in the virtual local area network to which the received Ethernet broadcast message belongs;

and step 107, packaging the received Ethernet broadcast message into an RPR broadcast message and forwarding the RPR broadcast message in an RPR ring network.

The application has the advantages that by aggregating the RPR nodes of the RPR ring network into virtual one DRNI virtual device, the device-level redundancy protection and the flow load sharing can be provided, the site protection of the RPR device can be realized, and the service load sharing among the RPR devices can be realized.

Fig. 2 is a schematic diagram of an RPR ring network with a distributed elastic network interconnection architecture provided by the present application, where node a and node B are aggregated and virtualized into a device through DRNI (Distributed Resilient Network Interconnect, distributed elastic network interconnection) to implement cross-device link aggregation, thereby providing device-level redundancy protection and traffic load sharing. The node a and the node B are distributed aggregation (Distributed Relay, DR) RPR nodes on the RPR ring network, and the node C and the node D are non-distributed aggregation RPR nodes of the RPR ring network.

The ethernet Port on the switch chip S1 of the node a and the ethernet Port on the switch chip S2 of the node B serve as IPPs (Intra-Port ports), and are connected through a physical ethernet Link serving as IPL (Intra-Port Link), and forward DRCP messages and data messages, synchronize MAC address entries and ARP entries, so that it is not necessary to synchronize all information of each member device like IRF (INTELLIGENT RESILIENT frame)/stacking system, and therefore the degree of coupling on the control plane is much smaller than stacking.

Besides the IPL link, there is also a KEEP ALIVE (keep-alive) link (not shown in fig. 2) between the two devices of node a and node B, which is used to detect the state of the peer distributed aggregation RPR node, i.e. to exchange KEEP ALIVE messages to perform dual-primary detection when the IPL link fails.

The port U1 of the node a and the port U2 of the node B are DR (Distributed RELAY INTERFACE) interfaces, which are connected to the same network device E outside the RPR ring network and belong to the same DR group (Distributed-Relay group) 210, the node a is a Primary Distributed aggregation RPR node device (Primary), and the node B is a standby Distributed aggregation RPR node (Secondary).

In fig. 2, when node a and node B send an ATD (Attribute Discovery ) frame in the topology discovery process of RPR, the RPP MAC address of the device is carried by using the newly added T-L-V field of the ATD frame as the notification field carrying the address of the DRNI member device.

The Internal physical ethernet Port between the switching unit S1 of the node a and the RPR processing unit R1 is an Internal Port1, and the Internal Port1 is mapped to Virtual ports (Virtual ports) 11, 12, 13, 14, 15 corresponding to the node B, C, D, the broadcast service, and the IPP Port in the RPR network 200, respectively.

The Internal physical ethernet Port between the switching unit S2 of the node B and the RPR processing unit R2 is an Internal Port2, and the Internal Port2 is mapped to virtual ports 21, 22, 23, 24, 25 corresponding to the node A, C, D, broadcast service, and IPP Port in the RPR network 200, respectively.

The Internal physical ethernet Port between the switching unit S3 of the node C and the RPR processing unit R3 is an Internal Port3, and the Internal Port3 is mapped to virtual ports 31, 32, 33, 34, and corresponds to the node A, B, D and the broadcast service in the RPR network 200, respectively.

The Internal physical ethernet Port between the switching unit S4 of the node D and the RPR processing unit R4 is an Internal Port4, and the Internal Port4 is mapped to virtual ports 41, 42, 43, 44, and corresponds to the node A, B, C and broadcast traffic in the RPR network 200, respectively.

The node a sets the virtual port 12 allocated to the node C by the device and the virtual port 22 allocated to the node C by the node B as DR group 1 (not shown in the figure), records the virtual port 12 and the virtual port 15 in the member port table entry of DR group 1, and sets the virtual port 13 allocated to the node D by the device and the virtual port 23 allocated to the node C by the node B as DR group 2, records the virtual port 13 and the virtual port 15 in the member port table entry of DR group 2. Node a sets a broadcast message filtering ACL table entry, wherein the matching entry is that the ingress port is IPP1 and the message type is a broadcast message, and the actions are to prohibit forwarding through the local member ports of DR group 210, DR group 1, and DR group 2.

The node B sets the virtual port 12 allocated to the node C by the equipment and the virtual port 22 allocated to the node C2 by the node A as DR group 1, records the virtual port 22 and the virtual port 25 in the member port table entry of the DR group 1, sets the virtual port 23 allocated to the node D by the equipment and the virtual port 22 allocated to the node D by the node A as DR group 2, and records the virtual port 23 and the virtual port 25 in the member port table entry of the DR group 2. The node B sets a broadcast message filtering ACL table entry, wherein the matching entry is that the ingress port is IPP2 and the message type is a broadcast message, and the actions are to prohibit the forwarding through the local member ports of the DR group 210, the DR group 1 and the DR group 2.

The node C sets the virtual ports 31 and 32 allocated to the node a and the node B as VP LAG3, and records the virtual port 31 and the virtual port 32 in the member port table entry of VP LAG 3.

Node D sets the virtual ports 41 and 42 allocated to node a and node B as VP LAG4, and records the virtual port 41 and the virtual port 42 as member ports in the member entries of VP LAG 4.

In the application, two equipment nodes A and B of the DRNI system are respectively provided with three DR groups, the DR group 210 is a distributed aggregation (DR) group for connecting with a network equipment E, and the DR group 1 and the DR group 2 are elastic packet ring distributed aggregation (RPR DR) groups for connecting with a node C and a node D on an RPR ring.

Fig. 3 is a schematic diagram of a node device forming a DRNI system forwarding an uplink broadcast message and a downlink broadcast in the ring network shown in fig. 2.

In fig. 3, node D receives the ethernet broadcast packet 301 sent by the terminal T1 of VLAN1000, and the switching unit S4 encapsulates the ethernet broadcast packet 301 with the RPR tag (e.g., 0 xff) of the broadcast service identifier, and sends the ethernet broadcast packet to the RPR processing unit R4 through the internet Port 4. The RPR processing unit R4 strips off the RPR tag, encapsulates the ethernet broadcast packet 301 into an RPR broadcast packet 302 with the RPR MAC address of the node as the source RPR MAC address and the RPR broadcast MAC address as the destination RPR MAC address, and forwards the packet on the inner ring and the outer ring of the RPR network 200.

The RPR processing unit R3 of the node C receives the RPR broadcast message 302, copies a part of the RPR broadcast message 302 to send to the node B, strips off the RPR message header of the received RPR broadcast message 302, adds the RPR tag with the identifier of the upper ring node D to the ethernet broadcast message 301 based on the source RPRMAC address of the RPR broadcast message 302, and sends the ethernet message 301 with the RPR tag with the identifier of the upper ring node D to the switching unit S3 through the international Port 3. The switching unit S3 of the node C strips off the RPR tag of the ethernet packet received by the internet Port3 according to the locally configured ACL entry, determines the virtual Port 33 corresponding to the upper ring node identifier D, learns the MAC address entry based on the source MAC address MAC T1 and the virtual Port 33 of the node D, and broadcasts the ethernet broadcast packet 301 in the VLAN 1000.

The RPR processing unit R2 of the node B receives the RPR broadcast message 302, reads the own device role of the DRNI system, determines that the own device role has low priority, and discards the received RPR broadcast message 302.

The RPR processing unit R1 of the node a receives the RPR broadcast message 302, determines that the role priority of the device is high, identifies the upper ring node as the node D according to the source RPRMAC of the RPR broadcast message 302, strips off the RPR message header of the RPR broadcast message 302, adds the RPR tag with the identifier of the upper ring node D to the ethernet broadcast message 301 of the inner layer, and sends the ethernet broadcast message 301 of the inner layer to which the RPR tag is added to the switching unit S1 through the Internal Port 1.

And the switching unit S1 of the node A strips off the RPR label of the inner Ethernet broadcast message 301 received by the Internet Port1 according to the locally configured ACL table entry, and determines the virtual Port 13 corresponding to the upper ring node identifier D. The switching unit S1 of node a learns the MAC address entries based on the source MAC address MAC T1 and DR set 2 of the upper ring node D.

The switching unit S1 of the node a sends an ethernet broadcast message 301 of the inner layer through the DR210 and the IPP port 1, and broadcasts the ethernet broadcast message 301 in the VLAN1000 (not shown in the figure), that is, the switching unit S1 of the node a copies one of the aggregate port of the DR group 210, the IPP port IPP1, and each ethernet port of the VLAN 1000. The node a selects the local DR interface U1 of the DR set 210 to send an inner ethernet broadcast message 301 to the network device E, sends an inner ethernet broadcast message 301 to the node B via IPP1, and sends an inner ethernet broadcast message 301 via each ethernet port within the VLAN1000 (not shown in fig. 3).

The switching unit S2 of the node B receives the ethernet broadcast message 301 through the IPP2 connected to the IPL link, and copies the ethernet broadcast message to the local DR interface U2 of the DR group 210, the local DR interface of the DR group 1, the local DR interface of the DR group 2, and each ethernet port of the VLAN 1000. The node B filters the ACL entry according to the set broadcast message, discards the ethernet broadcast message 301 copied for the local member port of the DR group 210, the local member port of the DR group 1, and the local member port of VPLAG, and broadcasts the ethernet broadcast message in the VLAN 1000.

Node a sends DRCP (Distributed Relay Control Protocol, distributed aggregation control protocol) message of the synchronous learning MAC address entry through IPP port 1 of the IPL link, where the outgoing interface in the MAC T1 entry is VP LAP2. The node B records the MAC T1 entry in the local MAC address entry.

In fig. 3, the upstream network device E receives the ethernet broadcast message 311 from the terminal T2, selects a member port connected to a member link of the node B in the link aggregation group (Link Aggregation Group, LAG) connected to the DRNI system, and sends the ethernet broadcast message 311 to the node B through the selected member port. The switching unit S2 of the node B receives the ethernet broadcast message 311, and learns the MAC address table entry according to the source MAC address MAC T2 and the aggregation port of the DR set 210.

The node B sends an Ethernet broadcast message 311 to the opposite node A through the IPP Port 2, broadcasts the Ethernet broadcast message 311 in the VLAN1000 (not shown in the figure), broadcasts an RPR broadcast message 312 on the RPR ring network, namely, a switching unit S2 of the node B is the IPP Port IPP2, each Ethernet Port of the VLAN1000 is duplicated, the Ethernet broadcast message 311 is sent through each Ethernet Port in the IPP Port IPP2 and the VLAN1000, a switching unit S2 of the node B packages an RPR label with a broadcast service identifier for the Ethernet broadcast message 311, and sends the Ethernet broadcast message to an RPR processing unit R2 through the Internal Port 2, the RPR processing unit R2 strips the RPR label, packages the Ethernet broadcast message 311 into an RPR broadcast message 312 with the RPR MAC address of the local node as a source RPR MAC address and the RPR MAC address as a destination RPR MAC address, and forwards the Ethernet broadcast message 311 on the inner ring and the outer ring of the RPR network 200.

The RPR processing unit R3 of the node C receives the RPR broadcast message 312, copies a part of the RPR broadcast message 312 and sends it to the node D, peels off the RPR message header of the received RPR broadcast message 312, adds the RPR tag with the identifier of the upper ring node B to the ethernet broadcast message 311 based on the source RPRMAC address of the RPR broadcast message 312 as the peeling-off message, and sends the ethernet message 311 with the RPR tag written in the identifier of the upper ring node B to the switching unit S3 through the Internal Port 3. The switching unit S3 of the node C strips off the RPR tag of the ethernet packet received by the internet Port3 according to the locally configured ACL entry, determines the virtual Port 32 corresponding to the upper ring node identifier B, learns the MAC address entry based on the source MAC address MAC T2 and the virtual Port 32, and broadcasts in the VLAN 1000.

The RPR processing unit R4 of the node D receives the RPR broadcast message 312, copies a part of the RPR broadcast message 312 and sends it to the node a, peels off the RPR message header of the received RPR broadcast message 312, adds an RPR tag with an identifier of the upper ring node B to the ethernet broadcast message 311 from which the message is peeled off, and sends the ethernet message 311 with the RPR tag written in the identifier of the upper ring node B to the switching unit S4 through the internet Port 4. The switching unit S4 of the node D strips off the RPR tag of the ethernet packet received by the internet Port4 according to the locally configured ACL entry, determines the virtual Port 42 corresponding to the upper ring node identifier B, learns the MAC address entry based on the source MAC address MAC T2 and the virtual Port 42, and broadcasts in the VLAN 1000.

When the RPR processing unit R1 of the node a receives the RPR broadcast message 312, determines that the role of the master device of the device is high, and identifies that the source RPRMAC address of the RPR broadcast message 312 is determined to be the DR node B, the received RPR broadcast message 312 is discarded.

The node B sends a DRCP message for synchronously learning the MAC address entry through the IPP port 2 of the IPL link, where the output interface in the MAC T2 entry is the DR group 120. The node a records a MAC T2 entry with an interface DR set 120 in the local MAC address entry.

Fig. 4 is a schematic diagram of a node device forming a DRNI system forwarding an uplink unicast message and a downlink unicast message in the ring network shown in fig. 2.

The node D receives the ethernet unicast message 401 sent by the terminal T1 to the terminal T2, the switching unit S4 of the node D finds that the output interface of the MAC address table entry matched with the destination MAC address MAC T2 of the ethernet unicast message 401 is VP LAG4, selects the member virtual port 41 in the member table entry of VP LAG4, and adds an RPR Tag (Tag) with a lower ring node identifier for the ethernet unicast message 401, where the lower ring node identifier is an identifier of the node a corresponding to the virtual port 41.

The switching unit S4 of the node D sends the ethernet unicast message 401 with the RPR tag to the RPR processing unit R4 through the internet Port 4. The RPR processing unit R4 of the node D strips off the RPR tag, encapsulates the RPR unicast packet 402 with the RPR MAC address D of the node as the source RPR MAC address, and the RPR MAC address a corresponding to the following ring node identifier as the destination RPR MAC address, and sends the packet to the node a through the RPR network 200.

The RPR processing unit R1 of the node a receives the RPR unicast message 402, peels off the RPR message header, encapsulates the RPR label with the upper ring node identifier for the ethernet message, that is, the upper ring node identifier corresponding to the RPR MAC address of the node D, and sends the upper ring node identifier to the switching unit S1 through the Internal Port international Port 1. The exchange unit S1 of the node a strips the RPR tag of the ethernet unicast packet 401 received by the international Port 1 according to the ACL entry, determines the corresponding DR group 2 according to the node D corresponding to the upper ring node identifier, searches the learned MAC address entry according to the source MAC address MAC T1 and the DR group 2, refreshes the aging time, searches the MAC entry of the destination MAC address MAC T2 in the MAC table, determines that the interface is an aggregation Port of the DR group 210, and sends the interface to the upstream network device through the local DR interface U1 of the aggregation Port of the DR group 210.

The network device E receives the ethernet unicast message 401, and sends the ethernet unicast message 401 to the terminal T2 according to the output port of the MAC address table entry matched with the destination MAC address.

When the network device E receives the ethernet unicast message 411 sent by the terminal T2 to the terminal T1, a member port connected to the node B is selected from the aggregation ports of the link aggregation group connected to the DRNI system, and the ethernet unicast message 411 is sent to the node B.

The switching unit S2 of the node B searches the outgoing interface DR group 2 of the MAC address table entry matched with the destination MAC address MAC T2 of the received ethernet unicast message 411, selects the local member virtual port 23 in the member table entry of DR group 2, and adds an RPR Tag (Tag) with a lower ring node identifier for the ethernet unicast message 411, where the lower ring node identifier is an identifier of the node D corresponding to the virtual port 23.

The switching unit S2 of the node B sends the ethernet unicast message 411 with the RPR tag to the RPR processing unit R2 via the internet Port 2. The RPR processing unit R2 of the node B strips the RPR tag, takes the RPR MAC address B of the node as the source RPR MAC address, takes the RPR MAC address D corresponding to the following ring node identifier as the destination RPR MAC address, encapsulates the ethernet unicast message 411 into the RPR unicast message 412, and sends the RPR unicast message 412 to the node C through the RPR network 200.

The node C transmits the RPR unicast message 412 to the node D according to the destination RPR MAC address of the RPR unicast message 412.

The node D receives the RPR unicast message 412, the RPR processing unit R4 strips off the RPR message header, encapsulates the ethernet unicast message 411 with the identifier of the upper ring node B based on the source RPR MAC address. The RPR switching unit R4 of the node D sends the ethernet unicast message 411 with the RPR tag to the switching unit S4 through the Internal Port international Port 4. The switching unit S4 strips the RPR tag of the ethernet unicast message 411 received by the international Port 4 according to the ACL entry, determines the corresponding VP LAG 4 according to the identifier of the upper ring node B, learns the MAC address entry of the MAC T1 according to the source MAC address MAC T2 and VP LAG 4 or refreshes the aging time of the learned MAC address entry of the MAC T1, searches out the Port of which the interface is the connection terminal T1 according to the destination MAC address of the ethernet unicast message 411, and sends the ethernet unicast message 401 to the terminal T1.

Fig. 5 is a schematic diagram of an embodiment of a packet forwarding device according to the present application, where the device at least includes a switching unit, an elastic packet ring processing unit, a CPU, and a memory. The switching unit may be implemented by a switching chip, and the resilient packet ring processing unit may be implemented by an FPGA chip. The processor is configured to execute the setup module by executing processor-executable instructions in the memory.

The setting module is used for setting an elastic packet ring distributed aggregation group of a non-distributed aggregation RPR node connected with the elastic packet RPR ring network, setting a distributed aggregation group of external network equipment connected with the RPR ring network, setting a broadcast message filtering table entry to prohibit forwarding of an Ethernet broadcast message received by an internal control link IPP port connected with a peer distributed aggregation RPR node through a local member port of the elastic packet ring distributed aggregation group and a local distributed aggregation port of the distributed aggregation group; the system comprises a switching unit, an elastic packet ring processing unit and a first MAC address table item, wherein the switching unit is used for receiving an Ethernet broadcast message through a distributed aggregation group, learning the first MAC address table item based on a source MAC address of the received Ethernet broadcast message and the distributed aggregation group, sending the received Ethernet broadcast message through an IPP port of an internal control link, broadcasting the received Ethernet broadcast message in a virtual local area network to which the received Ethernet broadcast message belongs, and the elastic packet ring processing unit is used for packaging the received Ethernet broadcast message into an RPR broadcast message and forwarding the RPR broadcast message in an RPR ring network.

The device comprises an RPR ring processing unit, a switching unit, a second MAC address table item, a virtual local area network and an IPP port, wherein the RPR ring processing unit is used for receiving RPR broadcast messages through the RPR ring network, discarding the received RPR broadcast messages when the device role is a standby distributed aggregation RPR node, identifying an upper link point of the received RPR broadcast messages when the device role is a main distributed aggregation RPR node, discarding the RPR broadcast messages when the upper link point of the received RPR broadcast messages is an opposite distributed aggregation RPR node, converting the received RPR broadcast messages into inner-layer Ethernet broadcast messages with an upper link point identification RPR label when the upper link point of the received RPR broadcast messages is a non-distributed aggregation RPR node, and the switching unit is used for learning the second MAC address table item according to a source MAC address of the inner-layer Ethernet broadcast messages and the distributed aggregation group of the elastic packet ring, and transmitting the inner-layer Ethernet broadcast messages through the distributed aggregation group and the IPP port.

The processing unit of the elastic packet ring is used for receiving RPR unicast message through the RPR ring network, identifying the upper ring node of the received RPR unicast message as a non-distributed aggregation RPR node, converting the received RPR unicast message into an inner Ethernet unicast message with an upper ring node identification RPR label, a switching unit used for learning a third MAC address table item according to the source MAC address of the inner Ethernet unicast message and the distributed aggregation group of the elastic packet ring, searching the matched MAC address table item of the destination MAC address of the inner Ethernet unicast message, determining that the output interface of the MAC address table item of the destination MAC address of the inner Ethernet unicast message is a distributed aggregation group, and transmitting the inner Ethernet unicast message through the local distributed aggregation interface of the distributed aggregation group.

The switching unit is used for receiving the Ethernet unicast message through a local distributed aggregation interface of the distributed aggregation group, learning a fourth MAC address table item based on the distributed aggregation group and a source MAC address of the received Ethernet unicast message, searching an MAC address table item matched with a destination MAC address of the received Ethernet unicast message, determining that an output interface of the MAC address table item of the destination MAC address of the received Ethernet unicast message is an elastic packet ring distributed aggregation group, the elastic packet ring processing unit is used for selecting a local member virtual port of the elastic packet ring distributed aggregation group, adding a lower link node identification RPR label with a non-distributed aggregation RPR node for the received Ethernet unicast message, stripping the RPR label of the received Ethernet unicast message, taking the RPR MAC address of the local node as the source RPR MAC address, taking the RPR MAC address corresponding to the non-distributed aggregation RPR node as the destination RPR MAC address, packaging the received Ethernet unicast message as the RPR unicast message, and transmitting the RPR unicast message through the RPR.

And the switching unit is used for sending an item synchronization message for synchronizing the first, second, third and fourth MAC address items through the IPP port so that the opposite-end distributed aggregation RPR node records the first MAC address item of which the interface is the distributed aggregation group of the elastic packet ring, records the second MAC address item of which the interface is the distributed aggregation group, records the third MAC address item of which the interface is the distributed aggregation group of the elastic packet ring and records the fourth MAC address item of which the interface is the distributed aggregation group.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the application.