WO2018001242A1

WO2018001242A1 - Data-message processing method and apparatus

Info

Publication number: WO2018001242A1
Application number: PCT/CN2017/090326
Authority: WO
Inventors: 张平平; 陈志伟; 孙军欢
Original assignee: 中兴通讯股份有限公司
Priority date: 2016-06-30
Filing date: 2017-06-27
Publication date: 2018-01-04
Also published as: CN107566237B; CN107566237A

Abstract

Provided in the embodiments of the present invention are a data-message processing method and apparatus; an SDN domain VLAN identifier and conventional domain VLAN identifier are preconfigured on a port; after a message transmitted by the port is received, a means for processing said message is determined according to VLAN information carried in the message and a VLAN identifier configured for the corresponding port. The means is such that the same port on a network device may simultaneously process an SDN domain service and a conventional domain service; it is unnecessary to set separate ports for an SDN domain service and a conventional domain service, and thus port resources of the network device are conserved and optimal resource configuration is achieved.

Description

Data message processing method and device

Technical field

The present invention relates to the field of communications devices, and in particular, to a data packet processing method and apparatus.

Background technique

After the network in the traditional Internet (Internet Technology) architecture is deployed according to business requirements, if the service requirements change, it is very cumbersome to re-configure the configuration on the corresponding network devices (routers, switches, firewalls). In the fast-changing business environment of the Internet/mobile Internet, the high stability and high performance of the network are not enough to meet the business needs. On the contrary, the flexibility and agility of the device configuration are more critical.

Therefore, SDN (Software Defined Network) emerges as the times require. SDN is a new type of network innovation architecture. It is an implementation of network virtualization. Its core technology OpenFlow OpenFlow through the control plane and data plane of network equipment. Separate and separate, thus achieving flexible control of network traffic, making the network more intelligent as a pipeline. What SDN does is to separate the control of the network device, managed by a centralized controller, without relying on the underlying network devices (routers, switches, firewalls), shielding the differences from the underlying network devices. At the same time, under the SDN architecture, the control of the network device is completely open, and the user can customize any network routing and transmission rule policies that he wants to implement according to his own expectations, which makes the network more flexible and intelligent.

Although the advantages of SDN network architecture are obvious, and the future development trend of the network is more biased towards SDN, in general, the application of traditional domain services is still widely used, and SDN has not completely replaced the traditional IT network architecture. Therefore, during the transition from the traditional IT network architecture to the SDN network architecture, a network device must be able to process the services of the traditional domain while having the SDN forwarding capability. Thus, dual mode forwarding of the traditional service plane and the SDN service plane is implemented on a network device.

In the prior art, the approach taken to achieve the effect of dual mode forwarding is: at SDN The switch is planned in advance, and it is planned which ports are SDN ports for processing SDN domain services, and which ports are traditional ports for handling traditional domain services. Once the planning is completed, the service that can be carried by a port has been determined: the SDN port can only process the SDN domain service. When the SDN port receives the unknown unicast packet, it will transmit the packet to the CPU. (Central Processing Unit), the CPU encapsulates the packet and sends it to the controller, and the controller determines the forwarding processing rule of the packet. A traditional port can only be used for traditional domain services. If a traditional port of an SDN switch receives a data packet, it will flood the data packet.

Although the above solution enables a switch to simultaneously handle the services of the SDN domain and the services of the traditional domain, the practice is simply to physically integrate the SDN switch with the traditional switch, and the SDN domain service and the traditional domain service are still performed through the port. Isolation is not conducive to the optimal configuration of resources, and the flexibility of network applications is not enough.

Summary of the invention

The data packet processing method and device provided by the embodiment of the present invention mainly solve the technical problem: when the dual-mode service processing of the network device is implemented in the prior art, the network device of the SDN domain and the traditional network are simply The ports of the device are physically fused. The ports of the network device cannot be reused, resulting in low resource utilization and low network flexibility.

To solve the above technical problem, an embodiment of the present invention provides a packet processing method, including:

Receiving the packet transmitted by the port of the network device, the port is configured with the SDN domain VLAN identifier and the traditional domain VLAN identifier, and the packet carries the VLAN information.

The processing manner of the packet is determined according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and the packet is processed accordingly.

The embodiment of the invention further provides a message processing device, including:

a receiving module, configured to receive a packet transmitted by a port of the network device, where the port is configured with an SDN domain VLAN identifier and a traditional domain VLAN identifier, where the packet carries VLAN information;

The processing module is configured to determine a processing manner of the packet according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and perform corresponding processing on the packet.

The embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the message processing method of any of the foregoing.

The beneficial effects of the invention are:

According to the packet processing method and device and the computer storage medium provided by the embodiment of the present invention, the SDN domain VLAN identifier and the traditional domain VLAN identifier are configured on the port in advance, and after receiving the packet transmitted by the port, according to the packet carrying the packet The VLAN ID and the VLAN ID configured on the corresponding port determine how the packet is processed. In this way, the same port on the network device can process the service of the SDN domain and the service of the traditional domain at the same time, and the corresponding port is not separately set for the service of the SDN domain and the service of the traditional domain, thereby saving the network device. Port resources enable optimal configuration of resources.

DRAWINGS

FIG. 1 is a flowchart of a packet processing method according to Embodiment 1 of the present invention;

2 is a flowchart of processing a data packet of an unknown unicast SDN domain according to Embodiment 1 of the present invention;

3 is a flowchart of processing a data packet of an unknown unicast SDN domain according to Embodiment 1 of the present invention;

4 is a schematic structural diagram of a packet processing apparatus according to Embodiment 2 of the present invention;

FIG. 5 is a schematic structural diagram of a switch according to Embodiment 3 of the present invention;

FIG. 6 is a flowchart of processing a packet by a switch according to Embodiment 3 of the present invention;

FIG. 7 is a schematic diagram of an application scenario of a packet processing method according to Embodiment 3 of the present invention.

detailed description

The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

Embodiment 1:

In the prior art, when the effect of the dual mode forwarding is implemented, the services carried by the ports of the network device can be hardened in the planning phase, so that the port function of the network device is rigid and cannot provide flexible service services for the user. More importantly, the port that handles the SDN domain service is separated from the port that handles the traditional domain service. This method wastes a lot of port resources and is not conducive to optimal resource configuration. Based on this, this embodiment provides a packet processing method, as shown in FIG. 1:

S102. Receive a packet transmitted by a network device port.

The network device in this embodiment may be a switch or a router. The SDN domain VLAN ID and the traditional domain VLAN ID can be configured on the port of the network device. For example, the VLAN ID of the SDN domain configured for the port of the network device is VLAN 3 to VLAN 9. The VLAN ID of the traditional domain is VLAN 13 to VLAN 19. This indicates that the port can handle the SDN domain services of VLANs 3 to 9 and the traditional domain services of VLANs 13 to 19. It can be understood that the VLAN identifier of the SDN domain and the VLAN identifier of the legacy domain should be different under the same port.

The packet should contain at least the VLAN information of the packet. The VLAN information can be used to determine the processing mode of the packet according to the VLAN identifier configured on the port that receives the packet.

S104. Determine, according to the VLAN information carried in the packet and the VLAN identifier configured on the port that transmits the packet, the processing mode of the packet, and perform corresponding processing on the packet.

In the network device, the hardware forwarding table stores some packet forwarding rules corresponding to the destination information, where the destination information includes the destination IP address or the destination MAC address of the packet. Taking the destination information as an IP address as an example, for a packet whose destination information is "125.120.218.106", the forwarding rule stored in the hardware forwarding table is transmitted from port 4; taking the destination information as a MAC address as an example, If the destination MAC address is "00-01-6C-06-A6-29", the packet is output from port 7 of the network device. Therefore, after receiving a packet, the port of the network device can pass The packet destination information carried in the packet is queried in the hardware forwarding table. If the forwarding rule corresponding to the destination information of the packet is already recorded in the hardware forwarding table, the network device can directly follow the packet according to the hardware. The forwarding rules in the forwarding table are transmitted. For those packets that cannot be queried from the hardware forwarding table, that is, packets that do not record forwarding rules in the hardware forwarding table, we call it "unknown unicast."

In the working principle of the switch, when the switch starts, its hardware forwarding table is empty. Therefore, if host A sends a packet to host B through the switch, the switch cannot determine the host when receiving the packet. On which port B is located (because there is no MAC or IP address of host B in the hardware forwarding table), this packet is an unknown unicast packet.

For an unknown unicast packet, the traditional domain is processed by flooding the packet. Flooding is a data stream delivery technology used by switches and bridges to send data streams received by an interface from all interfaces except the interface.

Obtaining a virtual forwarding instance of the packet according to the VLAN information carried in the packet, the virtual forwarding instance corresponding to the VLAN information in the packet, and corresponding to the port configured with the VLAN identifier corresponding to the VLAN information. The correspondence between the virtual forwarding instance and the port can be one-to-one, but more often than one-to-many. For example, the VLAN information carried in a packet is VLAN 3. According to VLAN 3, a virtual forwarding instance A can be obtained. The virtual forwarding instance A corresponds to all the ports in the network device configured with the VLAN ID of VLAN 3. Therefore, the information of all the ports with the VLAN 3 identifier in the network device can be obtained through the virtual forwarding instance as a medium. For example, in a network device, port 1, port 7, and port 8 all have the VLAN ID of VLAN 3. Therefore, port 1, port 7, and port 8 are the ports associated with the VLAN information in the packet. As a flood of exports of messages.

The SDN domain processes the unknown unicast packets differently from the traditional ones. The SDN port usually encapsulates the packets by using the protocol of the SDN domain, and then transmits the packets to the controller. The controller processes the packets according to a series of algorithms. After the calculation, the forwarding processing rule for the message is determined. Then, after receiving the forwarding processing rule sent by the controller, the forwarding processing may be performed according to the forwarding processing rule. The rule forwards the packet accordingly.

In the prior art, each port of the network device processes only one type of service, either an SDN domain service or a traditional domain service, so each port only receives the related services of the type that can be processed by itself. Message. In this embodiment, the SDN domain VLAN identifier and the traditional domain VLAN identifier are configured for one port at the same time. Therefore, one port can carry two types of services, that is, the SDN domain and the traditional domain. Therefore, the port of the network device in this embodiment is used. The received packet may not be simply a packet of the SDN domain or a packet of the traditional domain. However, the processing of these two types of messages is very different. Therefore, in this embodiment, after receiving the packet transmitted by the port of the network device, it is necessary to determine the processing manner for the packet, and it should be understood that determining the processing manner for the packet is actually the packet. Types are distinguished.

First, it is assumed that the SDN domain VLAN of VLAN 3 to VLAN 9 is configured for a port of the network device, and the traditional domain VLAN of VLAN 13 to VLAN 19 is also configured. On the basis of this, the VLAN information carried in the packet received by the port is VLAN 1. The network device cannot process the packet. Therefore, the packet can be directly discarded.

If the VLAN information carried in the received packet is VLAN 7, the device further determines whether the packet is an unknown unicast data packet. Determining whether a packet is an unknown unicast data packet can be mainly divided into two processes:

1. Determine whether the message is a data message. This judgment process can be performed according to the protocol identifier and the access control list included in the message. Any packet carries a protocol identifier. If a packet is a data packet, it may carry a relatively common TCP protocol identifier or a UDP protocol identifier. If it is a protocol packet, the protocol it carries is The identity is no longer the normal TCP protocol identifier or UDP protocol identifier. A specific type of protocol identifier is stored in the ACL. After obtaining the protocol identifier carried in the packet, the ACL can be searched in the ACL to determine that the protocol identifier carried in the packet is in the ACL. does it exist. If the protocol identifier carried in the packet exists in the ACL, the packet is a protocol packet. The protocol packet processing mode is simple and can be directly controlled. The CPU sent to the network device is sent by the CPU to the protocol stack module in the network device for processing. If the corresponding protocol identifier does not exist in the ACL, the packet is a data packet.

2. Determine whether the message is an unknown unicast message. The hardware forwarding table has been introduced in the foregoing, and the forwarding rules of the messages sent to these destinations are stored in the hardware forwarding table in units of destination information. Therefore, when a packet is received, the destination information carried in the packet may be found in the hardware forwarding table according to the destination information carried in the packet. If yes, the packet is not unknown. The broadcast packet can be directly forwarded by the network device. If the destination information carried in the packet does not exist in the hardware forwarding table, it indicates that the network device does not currently know which specific packet should be forwarded to. On the port, therefore, the message belongs to an unknown unicast message.

It can be understood that, in this embodiment, there is no strict time limit for determining that the message is a data message and the process of determining that the message is an unknown unicast message. It can be determined sequentially or simultaneously.

After determining that the packet belongs to an unknown unicast data packet, the packet may be determined to belong to the SDN domain or the legacy domain according to the VLAN information in the packet and the VLAN identifier of the port transmitting the packet. The method may be that the VLAN information in the packet matches the VLAN identifier of the SDN domain in the packet. If the matching succeeds, the packet belongs to the SDN domain. If the matching is unsuccessful, the packet belongs to the traditional domain. . Similarly, the VLAN information of the packet can be matched with the traditional domain VLAN identifier of the port. If the match is successful, the packet belongs to the traditional domain. Otherwise, the packet belongs to the SDN domain.

For the packets that belong to the SDN domain, the SDN domain can process the unknown unicast data packets. For the processing flow of unknown unicast data packets in the SDN domain, see Figure 2:

S202: Encapsulate the packet by using a protocol of the SDN domain, and then transmit the packet to the controller.

Generally, after determining that the packet belongs to the unknown unicast data packet of the SDN domain, the packet is sent to the protocol stack module in the network device, and the SDN protocol stack in the protocol stack module is used to match the packet according to the SDN domain protocol. The encapsulation process is performed, and the upper layer protocol tag is added and transmitted to the controller. The SDN protocol mainly includes OpenFlow and the like.

S204. Receive a processing table for the packet sent by the controller.

After receiving the encapsulated message, the controller determines how the message should be sent to its destination according to a series of algorithms or processing rules. These forwarding rules are included in the processing table and delivered to the network device.

S206. Forward the packet according to the processing table and update the information included in the processing table to the hardware forwarding table.

After the processing table is obtained, the packet may be sent out according to the forwarding rule included in the processing table. For example, in the processing table, the packet indicating that the destination information is A should be sent out on the network device's port 3, according to the indication. The message is transmitted to port 3. For all the data packets whose destination information is A in the SDN domain, the data packet can be forwarded from port 3 according to the forwarding rule. Therefore, the processing table can be updated to facilitate subsequent network device processing of subsequent packets. In the hardware forwarding table, after the packet whose destination information is also A is reappeared in the subsequent process, the packet can be processed according to the forwarding rule corresponding to the destination information A.

For packets that belong to the traditional domain, the processing of unknown unicast data packets can be processed according to the traditional domain. For the processing of unknown unicast data packets, see Figure 3:

S302. Obtain a corresponding virtual forwarding instance according to the VLAN information included in the packet.

In a network device, for example, a switch, the switchable chip can obtain a virtual forwarding instance from the forwarding control module. If the VLAN identifier of each port is different in a switch, the virtual forwarding instance can be obtained only according to the packet. The included VLAN information is carried out. If the two ports of a switch are the SDN port and the traditional port, and the two ports are configured with the same VLAN ID, you can obtain the virtual forwarding instance based on the VLAN information in the packet and receive the packet. The identification information of the port is used to obtain the virtual forwarding instance. The forwarding control module determines whether the packet is received by the SDN port or received by the traditional port according to the identification information of the port, so that the packet is allocated virtual forwarding according to the actual situation. Example.

S304. Determine, according to the virtual forwarding instance, a port that is a flooding outlet of the packet.

The virtual forwarding instance corresponds to the VLAN information in the packet. It also corresponds to the port configured with the VLAN ID corresponding to the VLAN information. The mapping between the virtual forwarding instance and the port can be one-to-one, but more The situation is one-to-many.

In the processing mode of the unicast data packet, the SDN domain uses the protocol of the SDN domain to encapsulate the packet, and the process of encapsulation processing is generally performed by the CPU of the network device. Therefore, when the packet is determined When the MPLS domain belongs to the unknown unicast data packet, the forwarding control module can also obtain the corresponding virtual forwarding instance. The virtual forwarding instance indicates that the flooding exit of the packet is the CPU.

S306. Flooding the packet to determine each port.

In the above description, the present invention provides a scheme for configuring the same VLAN identifier for an SDN port and a legacy port. In this solution, the same VLAN is configured on different ports, which can effectively save VLAN resources and improve resource utilization. .

In the network device, the underlying switching chip will report the MAC information or routing information of the packet after receiving the message, and then report the learning message to the upper layer software for processing by the upper layer software, but for the SDN domain, It is not necessary to need such a learning message, and therefore, it can be discarded directly. Therefore, in this embodiment, after receiving the learning message, it may first determine whether the learning message is an SDN domain, and if so, discard it directly, and if not, hand the learning message to the upper layer software for processing.

In the packet processing method provided in this embodiment, the SDN domain VLAN identifier and the traditional domain VLAN identifier can be configured on one port at the same time, and the VLAN information carried in the packet and the corresponding port are received when the packet is received. The configured VLAN ID determines how the packet is processed. In this way, the same port on the network device can process the service of the SDN domain and the service of the traditional domain at the same time, and the corresponding port is not separately set for the service of the SDN domain and the service of the traditional domain, thereby saving the network device. The port resource realizes the optimal configuration of resources, which is beneficial to the improvement of network application flexibility.

Embodiment 2:

The present embodiment provides a message processing apparatus. Referring to FIG. 4, the message processing apparatus 40 shown in FIG. 4 includes a receiving module 402 and a processing module 404.

The receiving module 402 is configured to receive a message transmitted by the network device port.

The processing module 404 determines the processing mode of the packet according to the VLAN information carried in the packet and the VLAN identifier configured on the port for transmitting the packet, and performs corresponding processing on the packet.

In the network device, the hardware forwarding table stores some packet forwarding rules corresponding to the destination information, where the destination information includes the destination IP address or the destination MAC address of the packet. Taking the destination information as an IP address as an example, for a packet whose destination information is "125.120.218.106", the forwarding rule stored in the hardware forwarding table is transmitted from port 4; taking the destination information as a MAC address as an example, If the destination MAC address is "00-01-6C-06-A6-29", the packet is output from port 7 of the network device. Therefore, after receiving a packet, the port of the network device can query the destination information of the packet carried in the packet to the hardware forwarding table. If it is found that the destination of the packet has been recorded in the hardware forwarding table. The forwarding rule of the information, the network device can directly transmit the packet according to the forwarding rule in the hardware forwarding table. For those packets that cannot be queried from the hardware forwarding table, that is, packets that do not record forwarding rules in the hardware forwarding table, we call it "unknown unicast."

In the working principle of the switch, when the switch starts, its hardware forwarding table is empty, so If host A sends a packet to host B through the switch at this time, when the switch receives the packet, it cannot determine which port the host B is on (because there is no MAC or IP address of host B in the hardware forwarding table) , so this message is an unknown unicast message.

The SDN domain processes the unknown unicast packets differently from the traditional ones. The SDN port usually encapsulates the packets by using the protocol of the SDN domain, and then transmits the packets to the controller. The controller processes the packets according to a series of algorithms. After the calculation, the forwarding processing rule for the message is determined. Then, after receiving the forwarding processing rule sent by the controller, the packet may be forwarded according to the forwarding processing rule.

In the prior art, each port of the network device processes only one type of service, either an SDN domain service or a traditional domain service, so each port only receives the related services of the type that can be processed by itself. Message. In this embodiment, the SDN domain VLAN identifier and the traditional domain VLAN identifier are configured for one port at the same time. Therefore, one port can carry two types of services, that is, the SDN domain and the traditional domain. Therefore, the port of the network device in this embodiment is used. Connect The received packet may no longer be simply a message in the SDN domain or a message in the traditional domain. However, the processing of these two types of messages is very different. Therefore, in this embodiment, after receiving the packet transmitted by the port of the network device, it is necessary to determine the processing manner for the packet, and it should be understood that determining the processing manner for the packet is actually the packet. Types are distinguished.

1. Determine whether the message is a data message. This judgment process can be performed according to the protocol identifier and the access control list included in the message. Any packet carries a protocol identifier. If a packet is a data packet, it may carry a relatively common TCP protocol identifier or a UDP protocol identifier. If it is a protocol packet, the protocol it carries is The identity is no longer the normal TCP protocol identifier or UDP protocol identifier. A specific type of protocol identifier is stored in the ACL. After obtaining the protocol identifier carried in the packet, the ACL can be searched in the ACL to determine that the protocol identifier carried in the packet is in the ACL. does it exist. If the protocol identifier of the packet exists in the ACL, the packet is a protocol packet. The protocol packet is processed in a simple manner. You can control the CPU directly to send it to the CPU of the network device. The protocol is sent to the protocol stack module in the network device for processing. If the corresponding protocol identifier does not exist in the ACL, the packet is a data packet.

2. Determine whether the message is an unknown unicast message. The hardware forwarding table has been introduced in the foregoing, and the forwarding rules of the messages sent to these destinations are stored in the hardware forwarding table in units of destination information. Therefore, when receiving a message, it can also be based on the message. The destination information is carried in the hardware forwarding table to check whether the destination information carried in the packet exists. If it exists, the packet is not an unknown unicast packet, and can be directly forwarded by the network device; If the destination information of the packet does not exist in the hardware forwarding table, the network device does not know which port to forward the packet to. The packet belongs to the unknown unicast packet.

For the packets belonging to the SDN domain, the processing module 404 can process the processing manner of the unknown unicast data packet according to the SDN domain:

The processing module 404 encapsulates the packet by using the protocol of the SDN domain and transmits the packet to the controller.

Generally, after determining that the packet belongs to the unknown unicast data packet of the SDN domain, the processing module 404 sends the packet to the protocol stack module in the network device, and the SDN protocol stack in the protocol stack module is based on the SDN domain protocol. The packet is encapsulated, and the upper layer protocol tag is added to the controller. The SDN protocol mainly includes OpenFlow and the like.

After receiving the encapsulated message, the controller determines how the message should be sent to its destination according to a series of algorithms or processing rules. These forwarding rules are included in the processing table and delivered to the network device. The processing module 404 receives a processing table for the message sent by the controller.

The processing module 404 forwards the message according to the processing table and further includes information included in the processing table. New to hardware forwarding table.

After the processing module 404 obtains the processing table, the packet may be sent according to the forwarding rule included in the processing table. For example, if the packet indicating that the destination information is A in the processing table should be sent by the port 3 of the network device, Transfer the message to port 3 according to the instructions. For all the data packets whose destination information is A in the SDN domain, the data packet can be forwarded from port 3 according to the forwarding rule. Therefore, in order to facilitate subsequent network device processing of subsequent packets, the processing module 404 can The processing table is updated to the hardware forwarding table. After the packet whose destination information is also A is reappeared in the subsequent process, the packet can be processed according to the forwarding rule corresponding to the destination information A.

For the packets that belong to the traditional domain, the processing module 404 can process the processing manner of the unknown unicast data packet according to the traditional domain:

The processing module 404 obtains a corresponding virtual forwarding instance according to the VLAN information included in the packet.

The virtual forwarding instance corresponds to the VLAN information in the packet. It also corresponds to the port configured with the VLAN ID corresponding to the VLAN information. The mapping between the virtual forwarding instance and the port can be one-to-one, but more The situation is one-to-many. The processing module 404 determines, as the virtual forwarding instance, a port that is a flooding outlet of the packet.

In the SDN domain, the SDN domain protocol is used to encapsulate the packet processing, and the process of encapsulation processing is generally performed by the network device. The CPU performs the process. Therefore, when it is determined that the packet belongs to the SDN domain and the unknown unicast data packet, the forwarding control module may also obtain the corresponding virtual forwarding instance, but the virtual forwarding instance indicates the flooding of the packet. The exit is CPU.

Finally, processing module 404 floods the message to determine each port.

In the network device, the underlying switching chip will report the MAC information or routing information of the packet after receiving the message, and then report the learning message to the upper layer software for processing by the upper layer software, but for the SDN domain, It is not necessary to need such a learning message, and therefore, it can be discarded directly. Therefore, in this embodiment, after receiving the learning message, the processing module 404 may first determine whether the learning message is an SDN domain, and if so, discard it directly, and if not, hand the learning message to the upper layer software for processing. .

The packet processing apparatus 40 provided in this embodiment may be deployed on a switch or a router, where the receiving module 402 may be implemented by a switch chip in a switch or a router, and the processing module 404 may be replaced by a switch chip in a switch or a router. The CPU is implemented together. In this embodiment, the controller may be a physical device or an application running on a general purpose server.

The packet processing apparatus 40 of the present embodiment can simultaneously configure the SDN domain VLAN identifier and the traditional domain VLAN identifier on one port, and according to the VLAN information and the corresponding port carried in the packet when receiving the packet transmission packet. The VLAN ID configured below determines how the packet is processed. In this way, the same port on the network device can process the service of the SDN domain and the service of the traditional domain at the same time, and the corresponding port is not separately set for the service of the SDN domain and the service of the traditional domain, thereby saving the network device. The port resource realizes the optimal configuration of resources, which is beneficial to the improvement of network application flexibility.

Embodiment 3:

The message processing method and the second embodiment in the first embodiment are provided below with reference to specific examples. The packet processing device is described. The network device in this embodiment takes a switch as an example. However, those skilled in the art should understand that the network device may also be a router. FIG. 5 shows a switch in this embodiment. Schematic diagram of the structure:

The switch 5 includes a switch chip 51 and a CPU 52. The switch chip 51 receives the message transmitted by the external device through the port. The switch provides a configuration interface. The user can configure the configuration of the switch 5 from the configuration interface. For example, the user can configure the VLAN ID of the SDN domain and the VLAN ID of the traditional domain on one port 511 of the switch chip 5.

The following describes the processing of the packet by the switch in this embodiment with reference to FIG. 6:

S601. The port 511 receives the packet.

After receiving the packet, the switch chip 51 can extract the VLAN information contained in the packet.

S602. The switch chip 51 determines whether the extracted VLAN information exists in the configuration of the port 511.

If yes, execute S603, if not, execute S604.

S603. The switch chip 51 determines whether the message is a data message.

If the VLAN ID corresponding to the VLAN information carried in the packet exists on the port 511, the device can determine whether the packet is a data packet according to the protocol identifier and the access control list included in the packet. If yes, execute S605, otherwise, execute S606.

S604. The switch chip 51 discards the packet.

If the VLAN information carried in the packet is not configured on port 511, the switch cannot process the packet. Therefore, the packet can be directly discarded.

S605. The switch chip 51 determines whether the message is an unknown unicast message.

If the packet is a data packet, the device further determines whether the data packet is an unknown unicast packet according to the destination information carried in the hardware forwarding table and the packet. If yes, execute S607; otherwise, execute S608.

S606. The switch chip 51 sends the message to the CPU.

If the packet is a data packet, it indicates that the packet is a protocol packet. The protocol packet is processed in a simple manner and can be directly sent to the CPU of the network device.

S607. The switch chip 51 determines whether the packet belongs to the SDN domain according to the VLAN information in the packet and the VLAN identifier of the port that transmits the packet.

When the judgment is made, the VLAN information in the packet is matched with the VLAN ID of the SDN domain in the packet. If the matching succeeds, the packet belongs to the SDN domain. If the matching is unsuccessful, the packet is reported. The text belongs to the traditional domain.

S608. The switch chip 51 forwards the packet according to the hardware forwarding table.

If the switching chip 51 determines that the forwarding rule of the packet destination information has been recorded in the hardware forwarding table, the forwarding process may be directly performed according to the corresponding forwarding rule.

S609. The switch chip floods the packet.

If the result of the judgment indicates that the packet belongs to the SDN domain, the packet can be reported to the CPU. If the packet is a Layer 2 service, it is based on the ACL (Access Control List ACL). If it is a Layer 3 service, it is routed. If the result of the judgment is no, the packet belongs to the traditional domain, and the packet can be directly flooded according to the processing manner of the unknown unicast data packet.

S610: The CPU determines, according to the VLAN information in the packet and the VLAN identifier of the port that transmits the packet, whether the packet belongs to the SDN domain.

The packets reported to the CPU are not all unicast data packets of the SDN domain, and may be protocol packets of the traditional domain. Therefore, the CPU needs to judge before performing the encapsulation process. When the judgment result is yes, Then execute S611.

S611. The CPU encapsulates the packet by using a protocol of the SDN domain.

Common encapsulation protocols can be performed using the OpenFlow protocol.

S612. The CPU sends the message to the controller 6.

After receiving the encapsulated message, the controller 6 determines how the message should be sent to its destination according to a series of algorithms or processing rules. These forwarding rules will be included in the processing table It is sent to the forwarding control module in the CPU.

S613. Forward the packet according to the forwarding rule sent by the controller, and update the hardware forwarding table.

FIG. 7 is an application scenario of the packet processing method provided by the embodiment: the server 71 and the server 72 respectively access the network device 73 and the network device 74, and the connection ports of the

network devices

73 and 74 are SDN ports, and the server The forwarding of service traffic of 71 and 72 is controlled by the SDN control plane. At the same time, the

servers

71 and 72 access the storage network and also access through the SDN instance port, but this part of the traffic goes away from the traditional control plane and is not controlled by the SDN controller. Since the

servers

71 and 72 are accessed through a single network card, the connection ports of the network device 73 and the network device 74 and the

servers

71 and 72 must follow the traditional plane control plane from the SDN control plane.

The packet processing method and device provided in this embodiment can ensure that the same port is controlled by the traditional protocol layer and can be controlled by the controller, which saves port resources and greatly increases the flexibility of application of the SDN switch in some scenarios. Sex.

Obviously, those skilled in the art should understand that the modules or steps of the above embodiments of the present invention can be implemented by a general computing device, which can be concentrated on a single computing device or distributed among multiple computing devices. On the network, optionally, they may be implemented by program code executable by the computing device, such that they may be stored in a computer storage medium (ROM/RAM, disk, optical disk) by a computing device, and at some In some cases, the steps shown or described may be performed in an order different than that herein, or they may be separately fabricated into individual integrated circuit modules, or a plurality of modules or steps may be fabricated into a single integrated circuit module. . Therefore, the invention is not limited to any particular combination of hardware and software.

The above is a detailed description of the embodiments of the present invention in conjunction with the specific embodiments, and the specific embodiments of the present invention are not limited to the description. It will be apparent to those skilled in the art that the present invention may be made without departing from the spirit and scope of the invention.

Industrial applicability

As described above, the data packet processing method and apparatus provided by the embodiment of the present invention have the following beneficial effects: the same port on the network device can simultaneously process the service of the SDN domain and the service of the traditional domain, and need not be separately The services of the SDN domain and the services of the traditional domain are respectively set to corresponding ports, thereby saving port resources of the network device and realizing optimal resource configuration.

Claims

A packet processing method includes:

Receiving the packet transmitted by the port of the network device, the port is configured with the SDN domain VLAN identifier and the traditional domain VLAN identifier, and the packet carries the VLAN information.

The processing manner of the packet is determined according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and the packet is processed accordingly.
The packet processing method according to claim 1, wherein the determining the processing manner of the packet according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and the report The corresponding processing of the text includes:

Determining that the packet is an unknown unicast data packet;

Determining that the packet belongs to an SDN domain according to the VLAN information in the packet and the VLAN identifier of the port;

The packet is processed according to the processing manner of the data message that is unknown to the unicast unicast.
The packet processing method according to claim 2, wherein the processing of the processing manner of the data packet in which the packet is not unicast according to the SDN domain includes:

The packet is encapsulated and processed by the protocol of the SDN domain and then transmitted to the controller;

Receiving, by the controller, a processing table for the packet, where the processing table includes destination information of the packet and a packet forwarding rule corresponding to the destination information;

And forwarding, according to the processing table, the packet, and updating information included in the processing table to a hardware forwarding table, where the destination information carried in the packet exists in the hardware forwarding table, determining the The packet does not belong to an unknown unicast packet.
The packet processing method according to claim 1, wherein the determining the processing manner of the packet according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and the report The corresponding processing of the text includes:

Determining that the packet is an unknown unicast data packet;

Determining, according to the VLAN information in the packet and the VLAN identifier in the port, that the packet belongs to a traditional domain;

The packet is processed according to the processing manner of the data packet whose unicast is unknown in the traditional domain.
The packet processing method according to claim 4, wherein the processing of the processing manner of the data packet according to the traditional domain unknown unicast includes:

Acquiring a corresponding virtual forwarding instance according to the VLAN information included in the packet, where the virtual forwarding instance is configured to determine a flooding outlet of the packet according to the VLAN information, where the flooding outlet is configured with the VLAN The port of the VLAN ID corresponding to the information;

Flooding the message to each of the ports.
The packet processing method according to claim 5, wherein when the port of the legacy domain and the port of the SDN domain have the same VLAN identifier, the VLAN information included in the packet and the identifier of the port receiving the packet are The information is obtained by the corresponding virtual forwarding instance. The port identification information is used to indicate that the port that receives the packet is a traditional domain port or an SDN domain port.
The packet processing method according to any one of claims 2-6, wherein the determining that the message is an unknown unicast data message comprises:

Extracting a protocol identifier and destination information included in the packet, where the destination information includes any one of a MAC address and a route;

Finding the protocol identifier in the access control list;

Finding the destination information in a hardware forwarding table;

And if the protocol identifier is not included in the access control list, and the destination information is not included in the hardware forwarding table, determining that the packet is an unknown unicast data packet.
A message processing device comprising:

a receiving module, configured to receive a packet transmitted by a port of the network device, where the port is configured with an SDN domain VLAN identifier and a legacy domain VLAN identifier, where the packet carries With VLAN information;

The processing module is configured to determine a processing manner of the packet according to the VLAN information carried in the packet and the VLAN identifier configured on the port, and perform corresponding processing on the packet.
The packet processing apparatus according to claim 8, wherein the processing module is configured to determine that the message is an unknown unicast data message; and according to VLAN information in the packet and a VLAN under the port The identifier determines that the packet belongs to the SDN domain, and processes the packet according to the processing manner of the data packet whose SDN domain is unknown unicast.
The packet processing apparatus according to claim 8, wherein the processing module is configured to determine that the message is an unknown unicast data message; and according to VLAN information in the packet and a VLAN under the port The identifier determines that the packet belongs to the traditional domain; and processes the packet according to the processing manner of the data packet whose unicast is unknown in the traditional domain.
A storage medium, the storage medium comprising a stored program, wherein the program is executed to perform the method of any one of claims 1 to 7.