WO2007031006A1 - A virtual switching method which could be routed - Google Patents

Abstract

A virtual switching method which could be routed, includes: STEP 1, determining the processing priority level of switching and routing at a interface; STEP 2, waiting for the arrival of the data message of the equipment’s interface; STEP 3, performing the switching process according to the processing priority level which determined at the interface, the input virtual LAN identification and the input interface number of the message, or performing the identifying and processing according to the three ply route or protocol information in the message. Of which if the processing procedure of the message is failed, the message will not be discarded, but processed with the low priority.

Description

 Routable virtual exchange method

 The present invention relates to the field of data network communication technologies, and in particular, to a routable virtual switching method. ' Background technique

 The number of metropolitan area network users is increasing, and the network scale and traffic are also increasing. The development direction of the metropolitan area network is from flat and tree to flat, diversified and customized, and the networking tends to be more flexible. Operators are more concerned about the performance-price ratio. At the same time, the support for value-added services is becoming more and more important. The requirements for network equipment are no longer simple traffic access and aggregation. Metropolitan area network services are becoming more abundant, requiring high-end network equipment to have more powerful service processing capabilities, especially for the aggregation layer network equipment to handle service capabilities: rich Layer 2 functions, MPLS functions, VPN, IP multicast, CDN, network storage, broadband video, e-commerce, V0D, QOS, NAT, IPV6. The IP metropolitan area network has new requirements for controllable, service differentiation, service quality, network security, new service support capability, and scalability. A single Ethernet technology or routing technology cannot meet the requirements of the current metropolitan area network. The operator is not limited to the router and switch mode. Correspondingly, more and more networks use routing devices to form IP metropolitan area networks. How to provide private line services to large groups through IP networks composed of routers, how to provide Layer 2 services in Layer 3 networks, how to Providing different services for users in the network is one of the problems that need to be solved after the transformation of the metropolitan area network.

 In the application environment shown in Figure 1, the service access device connects to the user network and the metropolitan area network. To meet the different needs of the customer, you need to complete the access of Layer 2 services and Layer 3 service functions. In this network environment, service access equipment can usually use the following technologies: 1. Multi-protocol label switching Layer 2 and Layer 3 virtual private network (MPLS L2 L3 VPN) technology, respectively, to transparently transmit Layer 2 PPPoE packets. L3 user access problems; 2, three-layer switching technology; 3, virtual switching VS TCH technology.

In the first solution, the service access device needs to use the pseudowire or virtual private local area network switching (VPLS) technology in the MPLS L2 VPN to transparently transmit the Layer 2 packet for the PPPoE user, and utilize the MPLS L3 VPN ( RFC2547bis) technology provides three layers of users to connect to the carrier network, access public networks or internal networks located in other geographic locations. Both use the label stack to provide transparent transmission of user messages over the carrier network, using the MPLS label stack. Need to be in PE and PE Establish connections, use outer tags (called TUNNEL tags) to penetrate the carrier network, and use inner tags (called VC tags) to distinguish between different VPNs. Through this technology, services such as service differentiation and flow control can be performed. However, this technology has a big disadvantage: It requires high equipment and needs to add more equipment, which complicates the network structure, and the equipment cost and management cost are greatly enhanced. In addition, the MPLS label also increases the length of the packet and reduces the transmission efficiency of the network.

 In the second scheme, the service access device uses Layer 3 switching technology to divide PPPoE access users and Layer 3 access clients into different VLANs, and uses Layer 2 switching and Layer 3 routing functions to complete access. For small networks, access problems can be better solved, but when there are more and more access users, serious VLAN ID shortages will result. In IEEE802. 1Q, the VLAN ID is only 12 bits long. Thus, in the entire Layer 2 network, only 4096 global VLANs can be supported, which is very different from the network operation requirements. Moreover, the use of pure three-layer switching technology is inconsistent with the trend of using network devices to form IP and MPLS metropolitan area networks. On the one hand, the inherent broadcast traffic of the Layer 2 network is large and the security is relatively weak. The characteristics cannot be solved. In another aspect, to support more and more new technologies, the requirements for new services for Layer 3 switches are increasing. The current Layer 3 switching technology cannot meet the requirements.

 The third technology is to enable the virtual switch method on the physical port to establish a switch table in the virtual switch. The contents of the switch table include: ingress port, inbound VLAN ID, outbound port, and outbound VLAN ID. After the packet enters the virtual switch, it searches the switch table based on the inbound port and the inbound VLAN ID of the packet to obtain the outbound and outbound VLAN IDs. If the exchange table does not exchange entries, the message is discarded. This technology can solve the problem of PPPoE user access in the problem described in Figure 1. For the access of the three-tier large customer, an additional interface needs to be deployed between the service router and the user SWITCH for special access. In this way, although the general virtual switching technology can efficiently perform Layer 2 packet processing according to the way of VLAN ID and port switching, and can alleviate the problem of insufficient VLAN ID, it is necessary to increase the network structure similar to FIG. The investment in equipment and links also affects the simplicity and manageability of the network to a certain extent. At the same time, when the number of access clients increases, the requirements for service access devices will increase. Greatly improve.

 Invention disclosure

The technical problem to be solved by the present invention is to provide a routable virtual switching method, which overcomes the shortcomings of the existing V-SWITCH technology that cannot support virtual switching and three-layer routing simultaneously on one device interface at the network access edge, and solves V- Switch technology cannot connect to customers on the same interface at the same time. The problem of providing Layer 2 data exchange functions and rich Layer 3 services; and, by the present invention, operators can provide high-efficiency Layer 2 virtual switching functions for large customers on one interface, and provide powerful service processing capabilities. For example, MPLS, multicast, VPN, and IPV6 expansion capabilities, without adding more devices and access lines.

 In order to achieve the above object, the present invention provides a routable virtual switching method, which is characterized in that it comprises the following steps:

 Step 1, determining the priority of switching and routing processing on an interface;

 Step 2: Wait for the arrival of the device interface data packet;

 Step 3: According to the processing priority determined by the interface, perform the exchange processing according to the inbound virtual local area network identifier and the inbound interface number of the data packet, or/and identify and process according to the three layers of information carried in the data packet. Forwarding, where the data packet is not discarded after the priority process fails, but is performed by a low priority process.

 The above routable virtual switching method is characterized in that, in step 1, the priority determining method of the interface is:

 If the network is focused on switching, the priority of the interface is the first exchanged route. Otherwise, the priority of the interface is the first route and then the exchange.

 The foregoing routable virtual switching method is characterized in that, in step 2, the validity check of the packet is included, and the validity check of the packet may include a frame check code check, a packet layer structure check, and an inspection step. Also includes:

 Check the frame check code of the packet. If the frame check code is incorrect, discard the packet.

 Checking the Layer 2 structure of the packet based on the interface type and the packet encapsulation type, and checking the encapsulation format of the packet according to the encapsulation protocol, and discarding the packet according to the encapsulation protocol and not supporting the encapsulation format.

 For packets that are not bridable, three layers of processing are performed directly.

 The above-mentioned routable virtual switching method is characterized in that, in step 3, the exchange processing is based on the lookup exchange table of the virtual local area network identifier and the inbound interface number of the packet to obtain the virtual local area network identifier and the outbound interface processing. Process

 The Layer 3 processing refers to the process of packet decision, forwarding, and processing according to the information above the link layer in the network layer included in the packet. As a common Layer 3 processing, the Layer 3 processing includes searching for a route according to an IP address. And perform packet forwarding decision and processing according to the routing result;

The protocol processing also includes routing protocol processing. The above-mentioned routable virtual switching method is characterized in that the structure of the switching table should include at least an inbound interface, an inbound VLAN ID, an outbound interface, and an outbound VLAN ID, and the exchange table has been previously established through configuration.

 The above-mentioned routable virtual switching method is characterized in that, in step 3, the three-layer processing further includes various encapsulation processing for performing an MPLS network, and also includes service processing according to IP information and an interface, including according to IP. The policy routing process performed by the address also includes various processing processes for the information of the second layer or more of the packet.

 The above-mentioned routable virtual switching method is characterized in that, in step 3, the sequence of the switching processing and the three-layer processing is determined according to the priority obtained by the interface, and the high priority process first processes the packet, which is high. Packets that cannot be processed by the priority process are processed into low-quality fe-level processes for processing. If the packets cannot be processed, the packets are discarded.

 The above-mentioned routable virtual switching method is characterized in that, in step 3, if the priority of the interface is first exchanged, after the packet is entered, the exchange table is checked first, and if the exchange table does not indicate, the message is not performed. Discarding, but performing Layer 3 processing or protocol processing based on the information carried in the packet;

 If the priority of the interface is the first, the switch performs the Layer 3 processing or the protocol processing based on the information carried in the packet. If the packet cannot be processed, the packet is not discarded. Exchange processing.

 The above routable virtual switching method is characterized in that the three-layer processing or protocol processing process comprises the following steps:

 Compare the destination MAC address of the message;

 If the destination MAC address is a broadcast MAC address or a multicast MAC address, it is processed by various protocols. If there is no relevant protocol, it is exchanged according to the exchange table.

 The above routable virtual switching method is characterized in that the three-layer processing or protocol processing further includes the following steps:

 Compare the destination MAC address of the message;

 If the destination MAC address matches the MAC address of the interface, the Layer 3 information carried in the packet is processed in three layers.

 If the MAC addresses do not match, the exchange table is checked according to the inbound virtual local area network identifier and the inbound interface number.

The above routable virtual switching method is characterized in that the step 3 includes the following steps - Step S31, if the processing priority determined by the interface is the first exchanged routing mode, then go to step S32; otherwise, the processing priority determined by the interface is the first routing and then switching mode, and the process proceeds to step S35;

 Step S32: Search for the exchange table by using the inbound port number of the data packet and the inbound VLAN ID. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist and the search fails, proceed to step S33.

 Step S33: Perform a discriminating process of the destination MAC address. If the broadcast or multicast MAC address is a unicast address equal to the MAC address of the interface, proceed to the next step S34; otherwise, the packet is discarded.

 Step S34, performing further three-layer processing and protocol processing according to the packet type. If it is an IP packet, performing IP service processing, for example, performing route lookup and forwarding according to the destination IP address; if it is a protocol packet, performing local protocol If yes, go to step S38, and the packets that cannot be processed are discarded.

 Step S35, performing the determination process of the destination MAC address, if it is a broadcast or multicast MAC address or a unicast address equal to the MAC address of the interface, proceed to the next step S36, otherwise, go to step S37;

 Step S36: Perform further Layer 3 processing and protocol processing according to the packet type. If it is an IP packet, perform IP service processing; if it is a protocol packet, perform local protocol processing, and go to step S38; , proceeds to the next step S37;

 Step S37: Searching for the exchange table by using the inbound port number and the inbound VLAN ID of the data packet. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist, the search fails and the packet is discarded.

 Step S38: Obtain information about the interface, re-encapsulate the packet, and send the packet to the outbound interface. Step S39: The step ends.

 The method of the present invention can effectively solve the problem that the V-switch technology cannot perform multiple Layer 2 and Layer 3 service processing on one interface at the same time, and can not only perform high-efficiency Layer 2 exchange of packets, but also can be used in the same manner. An interface completes various IP routing services and MPLS services. Expanding the access capabilities of service access devices also enables operators to meet the growing variety of new business requirements of customers with fewer devices and investments.

(1) Simultaneously provide Layer 2 and rich Layer 3 services for large customers on one interface, which can provide PPPoE access services and other virtual link functions, and provide various other IP routing services. It can even provide various MPLS services including MPLS L2 VPN and MPLS L3 VPN.

(2) It retains the features of the route and the broadcast isolation. It also increases the efficient packet exchange according to the inbound interface + VLAN ID, which simplifies the process of exchanging packets and processes the packets more efficiently. Text

 (3) Compared with MPLS VPN, there is no need to add extra labels to the packets, which can have better network transmission efficiency;

 (4) The network is simple, easy to maintain and manage, saving more investment for network operators and saving more operating costs.

 The invention is described in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

 1 is a network diagram of requirements in the prior art;

 2 is a schematic flow chart of implementing the present invention;

 Figure 3 is a detailed flow chart of the present invention;

 4 is a network diagram of an application of the routable virtual switching method of the present invention. The best way to implement the invention

 In the present invention, the technical solution of the routable virtual switching method is:

 A: Determine the priority of switching and routing processing on an interface;

 B: Wait for the arrival of the device interface data packet;

 C: Check the validity of the data packet. If the packet does not meet the legality, the error data packet is discarded.

 D: According to the determined processing priority, the data is exchanged according to the incoming virtual local area network identifier and the inbound interface number of the packet, or is identified and processed according to the Layer 3 routing or protocol information carried in the packet. After the packet processing fails in the priority process, the packet is not discarded, but the low priority process is performed.

 Further, step D may include the following information:

 D1: The exchange process is a process of obtaining a virtual local area network identifier and an outbound interface according to the lookup exchange table of the virtual local area network identifier and the inbound interface number of the packet;

D2: Layer 3 processing refers to finding routes based on IP addresses and forwarding packets according to routing results. At the same time, the three-layer processing may also perform various processing of the MPLS network, including other services performed according to IP information and interfaces, such as NAT, multicast, QoS, etc.; protocol processing refers to configuration Routing protocol processing, such as LACP processing, ARP processing, etc.;

 D3: The sequence of the exchange processing and the three-layer processing is determined according to the priority obtained in the previous step. The high priority process processes the packet first, and the high priority process cannot process the packet, and then transfers to the low priority process for processing. , still can not be processed, the message is discarded;

 D4: If the packet is configured to be exchanged first, the packet is checked. If the packet is not displayed, the packet is not discarded. The packet is processed according to the information carried in the packet.

 D5: If the configuration is performed first, the packet is processed first, and then the packet is processed according to the information carried in the packet. If the packet cannot be processed, the packet is not discarded, but the exchange table is continuously checked and exchanged;

 D6: The routing process may include: comparing the destination MAC address of the packet, if the destination MAC is a broadcast MAC address or a multicast MAC address, performing various protocol processing, and if there is no relevant protocol, exchanging according to the exchange table;

 D7: The routing process may further include: comparing the destination MAC address of the packet, if the destination MAC matches the MAC address of the interface, performing three layers of processing according to the three layers of information carried in the packet. If the MAC addresses do not match, they are exchanged according to the exchange table established by the incoming virtual LAN ID and the inbound interface number of the packet.

 The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

 As shown in FIG. 2, the method of implementing the present invention includes the following steps:

 Step S21: Determine the priority of switching and routing processing on an interface, and the priority of the interface includes "first exchanged route" and "first route after exchange";

 The principle of priority selection is: whether the interface data traffic is mainly exchanged, or is mainly routed, that is, whether the network focuses on switching or on routing. If you are focusing on the exchange processing, select "First exchange after routing", otherwise select "First route after exchange".

 Step S22: Waiting for the arrival of the device interface data packet;

Step S23: Check the legality of the data packet. If the packet does not meet the legality, the data packet is discarded. The validity check of the packet includes the frame check code check and the packet layer structure check. include: Step S231: Checking the frame check code of the packet, if the frame check code is incorrect, discarding the packet; Step S232: Performing a packet layer structure check according to the interface type and the packet encapsulation type, and the interface type is not limited to Ethernet. The interface can include other interfaces such as an ATM interface and a P0S interface. Correspondingly, the encapsulation format of the packet is checked, and the PPP encapsulation of the packet, the BCP encapsulation, the ATM Routed encapsulation, the ATM Bridged encapsulation, the Ethernet II, the IEEE 802. 1Q encapsulation, etc., are discarded according to the encapsulation protocol, and the encapsulation format packet is not supported. ;

 Step S233: Perform routing and protocol processing directly on the packets that are not bridable.

 Step S24: According to the processing priority determined in step S21, the processing is performed according to the inbound virtual local area network identifier and the inbound interface number of the packet, or the third layer routing or protocol information carried in the packet is identified and processed. After the packet processing fails in the priority process, the packet is not discarded, but is processed in a low priority process.

 The exchange process is a process of obtaining a virtual local area network identifier and an outbound interface according to the lookup exchange table of the virtual local area network identifier and the inbound interface number of the packet.

 According to the entry of the virtual local area network identifier and the inbound interface number of the packet, the interface is selected and the virtual local area network identifier is exchanged.

 Here, the structure of the exchange table should at least include (incoming interface, inbound VLAN ID, outbound interface, outbound VLAN ID), and the exchange table has been established through configuration or other means.

 If the bridged packet does not contain a VLAN encapsulation, the VLAN ID can take the value 0 or other value that does not conflict with the legal VLAN ID. If the switch table finds a hit, it is obtained from the exchange table (outbound interface, outbound). VLAN ID), the packet is re-encapsulated according to the VLAN ID, and the packet is sent to the outbound interface to complete the exchange process.

 In addition, Layer 3 processing refers to finding routes based on IP addresses and performing packet forwarding decisions and processing based on the routing results. At the same time, the Layer 3 processing may also perform various encapsulation processing of the MPLS network, including other services performed according to IP information and interfaces, such as MPLS, NAT, multicast, QoS, etc.; protocol processing refers to configuration for routing protocol processing, For example, LACP processing, ARP processing, and the like.

 The specific content of step S24 is different according to the priority determined in step S21. If the mode is set to "first exchange after routing", the packet is first exchanged. When the exchange cannot be processed, it enters Layer 3 processing or performs protocol processing.

If the configuration is "first route re-switching" mode, the packet is preferentially routed and protocol processed. When it cannot be routed, or does not meet the routing conditions, or does not meet the protocol processing conditions, enter the exchange. Reason.

 Referring to FIG. 3, step S24 specifically includes:

 Step S31: If it is the "first exchange after routing" mode, go to step S32, otherwise, it is "first route after exchange" mode, and proceed to step S35;

 Step S32: Searching for the exchange table by using the inbound port number and the inbound VLAN ID of the data packet. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist and the search fails, proceed to step S33.

 Step S33: Perform the determination process of the destination MAC address. If the broadcast or multicast MAC address or the unicast address is equal to the MAC address of the interface, perform the next step. Otherwise, the packet is discarded.

 Step S34: Perform further Layer 3 processing and protocol processing according to the packet type. If the IP packet is an IP packet, perform IP service processing, for example, performing route lookup and forwarding according to the destination IP address, and performing local protocol processing if it is a protocol packet. Go to step S38, the packet that cannot be processed is discarded. Step S35: Perform the discriminating process of the destination MAC address. If it is a broadcast or multicast MAC address or a unicast address equal to the MAC address of the interface, proceed to the next step. S36 processing, otherwise, proceeds to step S37;

 Step S36: Perform further Layer 3 processing and protocol processing according to the packet type. If it is an IP packet, perform IP service processing, for example, performing route lookup and forwarding according to the destination IP address, and performing local protocol processing if it is a protocol packet. Go to step S38, the message that cannot be processed, go to the next step S37;

 Step S37: Searching for the exchange table by using the inbound port number and the inbound VLAN ID of the data packet. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist, the search fails and the packet is discarded.

 Step S38: Obtain information about the outbound interface, re-encapsulate the packet, and send the packet to the outbound interface. Step S39: End of the step

As shown in Figure 4, the network includes two types of users: PPPoE access users and IP users. Two types of users connect to the service access router through the same interface, and the service access router connects the BAS device and the public network. The service access router connects to the user through the interface INT-A, and connects to the PPPoE access user through VLAN-1, and the IP user connects through the Ethernet interface. The packet is encapsulated in Ethernetll, and the service access router connected to the IP user allocates an address. For IP_A. Service access router and BAS equipment The connected interface is INT_B, which assigns VLAN-2. The interface connecting the service access router to the public network is INT_C, and the assigned address is IPJ. Assume that the work P address of the site D to be accessed is IP-D.

 The routable virtual switching method of the present invention is used in a service router. The exchange table is created. The contents of the exchange table include entries: (INT—A, VLAN— 1, INT—B, VLAN—2), and the index of the entry is (INT—A, VLAN—1). The exchange table also includes entries: (INT—B, VLAN_2, INT—A, VLAN—1), and the index of the entry is (INT_B, VLAN—2). A routing information table is established. The service router has a public network route to be accessed. The route includes entries: (IP_D, INT_C), and the contents of the routing entry are respectively the destination IP address to be accessed, IP-D, and the outbound interface INT- (:.

 The priority can be configured as two modes: "first exchange after routing" and "first route after exchange". After different user packets enter the service router, they will have different processing modes according to different processing priorities. In the network structure in Figure 3, it is configured as "first exchange after routing".

 The process of packet forwarding in the "first exchange rerouting" mode using the present invention is described as follows. After an IP user enters the service router, the device checks the validity of the packet and discards the incorrect packet. Because it has been determined that it is the "first exchange after routing mode", the VLAN ID of the packet is taken. In this specific case, the incoming packet is in the ETHERNET II encapsulation format. According to the system default configuration, the VLAN ID = 0, and the incoming interface is INT. — A. Then look up the exchange table according to (incoming interface, incoming VLANID), that is, (INT_A, 0). If there is no such entry in the exchange table, the search fails and returns. Therefore, the message is not exchanged, but the subsequent route processing or protocol processing is continued.

 After the IP user packet is not exchanged, the discard operation is not performed and the subsequent processing is continued. First, the destination MAC address is compared. If it is a broadcast MAC address and a multicast MAC address, the service router performs local processing, for example, performing an ARP response operation, so that the IP user can learn the MAC address of the gateway, that is, the service router INT-A. MAC address of the interface. The destination MAC address of the subsequent IP packet will be the address of the service router INT_A interface. If the destination MAC address is a unicast address, the destination MAC address of the compare packet and the address of the inbound interface INT A are not equal, and are discarded. If the MAC addresses are equal, they enter the subsequent routes or protocol processing according to the packet type. In this embodiment, the destination MAC address of the packet to be routed is the same as the address of the interface 1^^, and the encapsulated IP packet. Therefore, this packet can be sent to and from the routing entry (IP_D, INT-C). Interface INT-C to access the public network.

After the PPPoE user packet enters the service router in Figure 4, the legality check is performed. The inbound interface and the inbound VLAN ID are (INT_A, VLA _1), and the switch table is searched. The W entry (INT_A, VLAN-1, INT_B, VLAN_2), therefore, obtains the outbound interface and outbound VLAN ID, and repackages the packet with the new VLAN ID, VLAN-2, and sends it to INT_B. The message arrives at the BAS device. Similarly, the message from the BAS responds from the interface INT_B, uses the virtual local area network to identify the VLAN-2, and the service router searches the exchange table according to (INT-B, VLAN-2) to obtain the entry content (INT-A, VLA one). 1) The message can be re-used with the VLAN ID of the encapsulated virtual LAN and sent to the outbound interface INT_A. The packet can be returned to the PPPoE user.

 It can be seen from the above embodiments that the present invention can distinguish different data flows under one interface, realize high-efficiency VLAN switching, support various rich routing services, routing protocols provided by routers, MPLS functions, and the like. Business functions are supported. It retains the high efficiency of VLAN switching, and also preserves the broadcast isolation and rich features of the router. Various routing services can be operated without being affected.

 The invention may, of course, be embodied in a variety of other embodiments without departing from the spirit and scope of the invention. Changes and modifications are intended to be included within the scope of the appended claims. Industrial applicability

 The network of the present invention is simple and easy to maintain. The method provided by the invention enables an operator to provide high-efficiency Layer 2 virtual switching functions for large customers on one interface, and also provides powerful service processing capabilities, such as expansion capabilities for MPLS, multicast, VPN, and IPV6. Without adding more equipment and access lines.

Claims

Claim A routable virtual switching method, comprising the steps of:  Step 1, determining the priority of switching and routing processing on an interface;  Step 2: Wait for the arrival of the device interface data packet;  Step 3: According to the processing priority determined by the interface, perform the exchange processing according to the inbound virtual local area network identifier and the inbound interface number of the data packet, and/or identify and process according to the three layers of information carried in the data packet. Forwarding, where the data packet is not discarded after the priority process fails, but is performed by a low priority process.  2. The routable virtual switching method according to claim 1, wherein in step 1, the priority determining method of the interface is:  If the network is focused on switching, the priority of the interface is the first exchanged route. Otherwise, the priority of the interface is the first route and then the exchange.  The routable virtual switching method according to claim 2, wherein in step 2, the validity check of the packet is included, and the validity check of the packet includes a frame check code check and a packet. The second floor structure inspection, the inspection steps include:  Check the frame check code of the packet. If the frame check code is incorrect, discard the packet.  Checking the Layer 2 structure of the packet based on the interface type and the packet encapsulation type, and checking the encapsulation format of the packet according to the encapsulation protocol, and discarding the packet according to the encapsulation protocol and not supporting the encapsulation format.  For packets that are not bridable, routing and protocol processing is performed directly.  The routable virtual switching method according to claim 3, wherein in the step 3, the exchange processing is based on a lookup exchange table of the incoming virtual local area network identifier and the inbound interface number of the packet to obtain a virtual The process of processing the local area network identifier and the outgoing interface;  The Layer 3 processing refers to the process of packet decision, forwarding, and processing according to the information above the link layer in the network layer included in the packet;  As a common three-layer process, the three-layer processing includes searching for a route according to an IP address, and performing packet forwarding, decision, and processing according to the routing result;  The three-layer processing also includes routing protocol processing. The routable virtual switching method according to claim 4, wherein the structure of the exchange table comprises at least an inbound interface, an inbound VLAN ID, an outbound interface, and an outbound VLAN ID, and the exchange table It has been established in advance.  The routable virtual switching method according to claim 4, wherein in the third step, the three-layer processing further includes performing various encapsulation processing on the MPLS network, and also includes performing, according to the IP information and the interface. The service processing includes policy routing processing based on the IP address, and includes various processing processes on the information of the second layer or more of the packet.  The routable virtual switching method according to claim 4 or 5 or 6, wherein in step 3, the order of the exchange processing and the three-layer processing is determined according to the priority obtained by the interface, and the priority is The high-process process first processes the packet. If the packet cannot be processed in the high-priority process, the packet is processed in the low-priority process. If the packet cannot be processed, the packet is discarded.  The routable virtual switching method according to claim 7, wherein in step 3, if the priority of the interface is first exchanged, after the message is entered, the exchange table is checked first, if the exchange is performed. The table does not indicate that the packet is not discarded, but the Layer 3 processing or protocol processing is performed according to the information carried in the packet.  If the priority of the interface is the first, the switch performs the Layer 3 processing or the protocol processing based on the information carried in the packet. If the packet cannot be processed, the packet is not discarded. Exchange processing.  9. The routable virtual switching method according to claim 7, wherein the three-layer processing or protocol processing comprises the following steps:  Compare the destination MAC address of the message;  If the destination MAC address is a broadcast MAC address or a multicast MAC address, it is processed by various protocols. If there is no relevant protocol, it is exchanged according to the exchange table.  The routable virtual switching method according to claim 9, wherein the three-layer processing or protocol processing further comprises the following steps:  Compare the destination MAC address of the message;  If the destination MAC address matches the MAC address of the interface, the Layer 3 information carried in the packet is processed in three layers.  If the MAC addresses do not match, the exchange table is checked according to the inbound virtual local area network identifier and the inbound interface number. 11. The routable virtual switching method according to claim 8, wherein the step 3 comprises the following steps: Step S31, if the processing priority determined by the interface is the first exchanged routing mode, then go to step S32; otherwise, the processing priority determined by the interface is the first routing and then switching mode, and the process proceeds to step S35;  Step S32: Search for the exchange table by using the inbound port number of the data packet and the inbound VLAN ID. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist, the search fails, and then step S33 is performed;  Step S33: Perform the discriminating process of the destination MAC address. If the broadcast or multicast MAC address or the unicast address is equal to the MAC address of the interface, proceed to the next step S34; otherwise, the packet is discarded;  Step S34, performing further three-layer processing and protocol processing according to the packet type. If the IP packet is an IP packet, performing IP service processing, for example, performing route lookup and forwarding according to the destination IP address; if it is a protocol packet, performing local protocol processing, Go to step S38, the unprocessable message is discarded; Step S35, the destination MAC address is determined, if it is a broadcast or multicast MAC address or a unicast address equal to the MAC address of the interface, proceed to the next step. S36 processing, otherwise, proceeds to step S37;  Step S36: Perform further Layer 3 processing and protocol processing according to the packet type. If it is an IP packet, perform IP service processing; if it is a protocol packet, perform local protocol processing, and go to step S38-, the report that cannot be processed. Text, proceeds to the next step S37;  Step S37: Searching for the exchange table by using the inbound port number and the inbound VLAN ID of the data packet. If the forwarding table is found successfully, go to step S38. If the lookup forwarding table entry does not exist, the search fails and the packet is discarded.  Step S38: Obtain information about the interface, re-encapsulate the packet, and send the packet to the outbound interface. Step S39, the step ends.