WO2009138036A1 - 以太网业务传送中转发表的更新方法、设备和系统 - Google Patents

以太网业务传送中转发表的更新方法、设备和系统 Download PDF

Info

Publication number
WO2009138036A1
WO2009138036A1 PCT/CN2009/071782 CN2009071782W WO2009138036A1 WO 2009138036 A1 WO2009138036 A1 WO 2009138036A1 CN 2009071782 W CN2009071782 W CN 2009071782W WO 2009138036 A1 WO2009138036 A1 WO 2009138036A1
Authority
WO
WIPO (PCT)
Prior art keywords
mac address
forwarding table
link
port
primary
Prior art date
Application number
PCT/CN2009/071782
Other languages
English (en)
French (fr)
Inventor
江元龙
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2009138036A1 publication Critical patent/WO2009138036A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/02Topology update or discovery
    • H04L45/021Ensuring consistency of routing table updates, e.g. by using epoch numbers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/02Topology update or discovery
    • H04L45/028Dynamic adaptation of the update intervals, e.g. event-triggered updates
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/02Topology update or discovery
    • H04L45/04Interdomain routing, e.g. hierarchical routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/22Alternate routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/28Routing or path finding of packets in data switching networks using route fault recovery
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/68Pseudowire emulation, e.g. IETF WG PWE3

Definitions

  • the present invention relates to the field of communications, and in particular, to a method, device and system for updating a forwarding table in an Ethernet service transmission.
  • Ethernet is currently the most widely deployed LAN technology, but it has some limitations, such as VLAN (Virtual Local Area Network), limited address space, limited spanning tree protocol scalability, and MAC (Media Access Control) address. It is not easy to aggregate and other defects.
  • VLAN Virtual Local Area Network
  • MAC Media Access Control
  • VPLS Virtual Private LAN Service
  • L2VPN Layer 2 Virtual Private Network
  • Virtual Private LAN Service Virtual Private LAN Service
  • IP/MPLS Multi-Protocol Label Switching
  • the network architecture reference model of the VPLS is used.
  • the customer's Ethernet service is connected to the PE (Provider Edge) device through the CE (Customer Edge) device.
  • One or more P devices (for example, P routers) are connected together.
  • a VPLS instance is composed of multiple PE devices.
  • the PE device receives an Ethernet frame from the CE device through the AC (Attach Circuit)
  • the VPLS forwarder device or module with a MAC address forwarding table
  • the PW (Pseudowire) transmitted on the MPLS network is obtained according to the information such as the destination MAC address in the Ethernet frame, and then the PW is multiplexed into the tunnel of the packet switching network to complete the transparent transmission of the Layer 2 packet.
  • PBBN For the PBBN network, a similar architecture can be used to complete the transmission of Ethernet multipoint services, but only the pseudo The line belongs to another logical channel, and the multiplexed network tunnel is the Ethernet switching path.
  • the above MAC address forwarding table is established through a MAC address self-learning mechanism, similar to the Ethernet self-learning process, which is briefly described as follows:
  • the network edge device When the network edge device (PE) is from an AC/PW (for convenience, the connection point of the AC or PW to the PE is called a port, it may be a physical port, or it may be just a logical port, such as the corresponding PW.
  • the packet When receiving a packet, if the destination MAC address of the packet is a broadcast address or an unknown address, the packet will be flooded to all other ports; if the source MAC address of the packet is unknown (refer to the MAC address forwarding table) If the entry corresponding to the entry is not found, the source MAC address is bound to the receiving port of the packet, and is stored in the MAC address forwarding table, and the new packet with the destination MAC address being the address is sent to the new packet. port. This process is also known as the MAC address self-learning process.
  • the binding relationship between the above MAC address and the port may become invalid.
  • the CE1 device is connected to the PE1 device through AC1.
  • the service packets are transmitted from PW1 to the PE3 device and finally to the client CE2 device.
  • AC1 fails, CE1 switches to AC2 to access the PE2 device.
  • the service is transmitted from PW2 to the PE3 device and finally to the customer CE2 device.
  • the MAC address forwarding table on the PE1 device, the PE2 device, and the PE3 device related to the foregoing switching operation needs to be updated to reflect the new topology relationship.
  • the following methods are provided for the MAC address. Update of the forwarding table:
  • the address aging mechanism if the MAC address in the MAC address forwarding table is not used within a certain period of time, the entry corresponding to the MAC address is deleted, and the new MAC address forwarding entry is obtained through the self-learning process. Update of the MAC address forwarding table.
  • the PE device When the network topology changes, the PE device sends a MAC address recovery message to all other peer PE devices in the VPLS instance.
  • the message contains a list of MAC addresses to be cleared. Other PE devices delete corresponding MAC address entries according to the message. .
  • the PE device sends an address reclaim message carrying an empty MAC address table, all MAC addresses associated with the VPLS instance are cleared, except for the MAC address learned from the PW receiving the message.
  • a TIN (Type-Length-Value) field carrying a PE device identifier is added to the signaling by extending the LDP (Label Distribution Protocol) signaling used by the VPLS.
  • LDP Label Distribution Protocol
  • a generic protocol field representation that specifies that only all MAC addresses learned from a particular PE device are cleared to greatly reduce the range of MAC address cleanup.
  • the inventors have found that the prior art has at least the following problems:
  • the above methods all need to clear the corresponding entries in the MAC address forwarding table, and then re-learn through the self-learning method to obtain a new MAC address forwarding table.
  • packet flooding will occur during the self-learning process, a large number of redundant messages will be generated in the core network, increasing network bandwidth consumption.
  • the update delay of the MAC address forwarding entry is large, which may result in the packet being sent to the correct PE device in the new topology, but sent to the PE device in the old topology. Summary of the invention
  • the embodiment of the present invention provides a method, device, and system for updating a forwarding table in an Ethernet service transmission.
  • the technical solution is as follows:
  • a method for updating a forwarding table in an Ethernet service transmission comprising:
  • the carrier edge (PE) device updates its own MAC address forwarding table based on the primary link, the alternate link, and the media access control (MAC) address list.
  • PE carrier edge
  • An operator edge device comprising:
  • the fault processing module (301) is configured to switch the primary link to the standby link when the link or the carrier edge (PE) device fails;
  • a forwarding table update module (302) configured to: when the fault processing module (301) switches from the primary link to the standby link, according to the primary link, the backup link, and a media access control (MAC) address list Update its own MAC address forwarding table.
  • MAC media access control
  • An operator edge device comprising:
  • a judging module configured to determine, when the primary link switches to the standby link, whether there is an end-operator edge (PE) device on the standby link as itself;
  • PE end-operator edge
  • a processing module (402) configured to: when the judgment result of the determining module (401) is that the standby link has a PE device at one end, and the self-media access control (MAC) address forwarding table The port corresponding to the entry of the primary link is replaced with the port corresponding to the standby link; otherwise, the MAC address forwarding table migration message is sent to the locally corresponding PE device on the standby link.
  • MAC self-media access control
  • An operator edge device comprising:
  • a receiving module configured to receive a message including a media access control (MAC) address list;
  • MAC media access control
  • the processing module (502) after receiving the message, the receiving module (501) updates its own MAC address forwarding table according to the MAC address included in the MAC address list, and sets the port of the MAC address. Is the port corresponding to the standby link.
  • An update system for forwarding tables in an Ethernet service transmission comprising:
  • the primary carrier edge device (601) is configured to determine whether the backup link has one end operator edge device (PE) as its own when the primary link is switched to the standby link; if yes, its own media access control (MAC: The port corresponding to the primary link corresponding to the primary link in the address forwarding table is replaced with a port corresponding to the standby link; otherwise, the MAC address forwarding table migration message is sent to the standby carrier edge device;
  • PE end operator edge device
  • the standby carrier edge device (602) is configured to: after receiving the MAC address forwarding table migration message of the primary carrier edge device (601), add the MAC address included in the MAC address forwarding table migration message to its own MAC address.
  • the port of the MAC address entry is set to a port corresponding to the standby link.
  • An update system for forwarding tables in an Ethernet service transmission comprising:
  • the determining device (701) is configured to: when the primary link is switched to the standby link, determine whether the switching of the primary link to the standby link causes a change of a carrier edge (PE) device, and if yes, to the primary device
  • the primary carrier edge device (702) on the link sends a media access control (MAC) address forwarding table migration notification; otherwise, sends a MAC address forwarding table switching notification to the primary carrier edge device (702) on the primary link.
  • MAC media access control
  • the primary carrier edge device (702) is configured to send a MAC address forwarding table migration to the corresponding alternate carrier edge device (703) on the standby link after receiving the MAC address forwarding table migration notification sent by the determining device. After receiving the MAC address forwarding table switching notification sent by the determining device (701), replacing the port corresponding to the primary link corresponding entry in the MAC address forwarding table with the standby link Corresponding port;
  • the standby carrier edge device (703) is configured to: after receiving the MAC address forwarding table migration message of the primary carrier edge device (702), add the MAC address included in the MAC address forwarding table migration message to its own MAC address.
  • the entry of the MAC address entry port is set to a port corresponding to the standby link.
  • the PE device on the standby link updates the port corresponding to the primary link in the MAC address forwarding table of the device to the port corresponding to the standby link, thereby avoiding the recovery of the MAC address.
  • the re-learning process and the packet flooding caused by it reduce the impact on the network service, improve the update efficiency of the MAC address forwarding table, and save the bandwidth of the core network.
  • FIG. 1 is a schematic diagram of a network architecture reference model of a VPLS provided by the prior art
  • FIG. 2 is a schematic structural diagram of a networking provided by the prior art
  • Embodiment 3 is a schematic structural diagram of a networking provided by Embodiment 1 of the present invention.
  • FIG. 4 is a flowchart of a method for updating a forwarding table in an Ethernet service transmission according to Embodiment 1 of the present invention
  • 5 is a schematic structural diagram of a networking provided by Embodiment 2 of the present invention
  • 6 is a flowchart of a method for updating a forwarding table in an Ethernet service transmission according to Embodiment 2 of the present invention
  • Embodiment 7 is a schematic structural diagram of another networking provided by Embodiment 2 of the present invention.
  • Embodiment 8 is a schematic structural diagram of a networking provided by Embodiment 3 of the present invention.
  • Embodiment 9 is a schematic structural diagram of a networking provided by Embodiment 4 of the present invention.
  • FIG. 10 is a schematic structural diagram of networking after the PW switching in FIG. 9 according to Embodiment 4 of the present invention.
  • FIG. 11 is a schematic structural diagram of a carrier edge device according to Embodiment 5 of the present invention.
  • FIG. 12 is a schematic structural diagram of a carrier edge device according to Embodiment 6 of the present invention.
  • FIG. 13 is a schematic structural diagram of a carrier edge device according to Embodiment 7 of the present invention.
  • FIG. 14 is a schematic structural diagram of an update system of a forwarding table in an Ethernet service transmission according to Embodiment 8 of the present invention
  • FIG. 15 is a schematic structural diagram of an update system of a forwarding table in an Ethernet service transmission according to Embodiment 9 of the present invention.
  • the PE device connected to the standby link updates the port corresponding to the primary link in the MAC address forwarding table of the device to the port corresponding to the standby link.
  • the process of MAC address recovery and self-learning is not required, and the delay of the MAC address forwarding table update process and the consumption of the network bandwidth are reduced.
  • the MAC address forwarding table provided by the embodiment of the present invention includes: a destination MAC address and port information, where the port can be either a physical port or a logical port.
  • the link in the embodiment of the present invention may be a physical link or a logical link, where there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port) as an example.
  • the technical solution provided by the embodiment of the present invention is still applicable to the case where the link is a physical link and the port is a physical port.
  • the logical port in the following embodiment of the present invention is represented by a PW.
  • VPLS deployments generally use a redundant network architecture.
  • Some access circuits AC, PW, and PE devices fail, they can continue to use standby AC, PW, or PE devices.
  • multiple redundant PWs can also form a set of load sharing so that Ethernet services can be simultaneously transmitted on these PWs.
  • the PEs and the PWs that are mutually redundant or load-sharing can form a protection group. Therefore, all PEs or PWs in a protection group form a redundant backup or load balancing relationship. Group relationships can be established using management configurations or It is realized by control signaling.
  • the management configuration can be manually set through the management platform, or it can be calculated through a dedicated routing platform such as PCE (Path Computation Element).
  • PCE Policy Computation Element
  • a PW is set up, a PW is configured for the PW, and the PW and the standby PW are associated with each other in the same protection group.
  • the PW protection group and its two ends associated with each VPLS instance are saved on the PE.
  • the binding relationship of the PE device identifier that is, the corresponding PE device can be learned by the PW. In this way, the data service can still be forwarded through the standby PW when the primary PW fails.
  • a fully-connected LDP session is established between all PEs.
  • the SP has an LDP session between each PE device to control signaling through these LDP sessions.
  • the main PW will switch to standby through the network management platform or manual commands, or in the event of the following failures.
  • the primary AC is faulty.
  • the fault detection of the AC can be detected by physical signal detection, APS (Automatic Protection Switching), LACP (Link Aggregation Control Protocol), 802.1ag CC (Continuity and Connectivity Check, continuity detection)
  • the main PE or the main PW is faulty.
  • the fault detection of the PE or the PW can be implemented through the routing protocol, VCCV (Virtual Circuit Connection Verification), and Bidirectional Forwarding Detection (BFD). .
  • VCCV Virtual Circuit Connection Verification
  • BFD Bidirectional Forwarding Detection
  • the service is transferred to the new AC, PW, or PE device to ensure that the service is as unaffected as possible.
  • some dual-homing protection protocols can perform state synchronization between PE devices to ensure that when a primary AC or primary PE device fails, the CE is automatically allowed to transmit traffic through the standby AC, the standby PW, or the standby PE device.
  • the dual-homing protection protocol is usually a private protocol.
  • the local PE device of the primary PW can be classified according to whether the standby PW passes the PE device, that is, whether the same PE device passes before and after the switchover.
  • the update process of the MAC address forwarding table brought by the PW switch can be divided into There are two cases of MAC address forwarding table migration (local PE device change) and MAC address forwarding table switching (local PE device unchanged). The details are described in detail in the following embodiments. Example 1
  • This embodiment provides a method for updating a forwarding table in an Ethernet service transmission.
  • PW1 and PW2 are a primary PW and a standby PW, respectively.
  • This embodiment takes place in PW1 in FIG. Fault, the service is switched from PW1 to PW2 as an example, because switching from PW1 to PW2 does not cause the local PE device to send
  • the change of the MAC address forwarding table is performed by using the MAC address forwarding table switching method in this embodiment.
  • the method for updating the MAC address forwarding table includes:
  • Step 101 After detecting that PW1 fails, PE1 and PE3 switch from PW1 to PW2, that is, stop using PW1 and activate PW2.
  • Step 102 PE1 and PE3 determine that the local PE device does not change before and after the PW switch, and respectively
  • the PW1 in the MAC address forwarding table is updated to PW2.
  • the general PW identifier will contain the identifier (such as the address) of its endpoint (PE), so as long as the detected endpoint identifier (address) of PW1 is compared with the PE node's own address, if the same , the local PE device does not change before and after the PW switch. If it is different, the description has changed.
  • PE endpoint
  • PE1 and PE3 first check the MAC address forwarding table of the device, and find out the corresponding entry of the port as PW1 from the MAC address forwarding table, and then replace PW1 in each entry with PW2.
  • PE1 and PE3 After the update of the MAC address forwarding table is completed, PE1 and PE3 transmit the data packet whose destination address is the MAC address through PW2.
  • PE1 may send a forwarding table update notification to PE3 after updating its own MAC address forwarding table.
  • the forwarding table update notification may be implemented in the form of a MAC_Switch (PW1, PW2, MAC_List) message, indicating that the port corresponding to the address in the MAC_List in the MAC address forwarding table is updated from PW1 to PW2, and the message may be encapsulated by the message format in the LDP. And communicating through the LDP session. If the MAC_List is empty, it indicates that all ports in the MAC address forwarding table are PW1 addresses;
  • PE3 After receiving the forwarding table update notification sent by PE1, PE3 is based on the content of the notification (for example, MAC_Switch(PWl,
  • PE1 and PE3 after switching from PW1 to PW2, PE1 and PE3 modify their own MAC address forwarding table, and PE1 and PE3 forward packets through PW2, which avoids the recovery and re-learning process of MAC address and packet flooding caused by it. , reducing the impact on the network service, improving the update efficiency of the MAC address forwarding table, and saving the bandwidth of the core network.
  • Example 2
  • This embodiment provides a method for updating the forwarding table in the Ethernet service transmission.
  • PE3 and PE4 are the primary PE device and the standby PE device respectively.
  • PW1 and PW2 are the primary PW and the PW.
  • the standby PW, PE1 is connected to PE3 and PE4 through PW1 and PW2 respectively.
  • PW1 in Figure 5 is faulty, and PW1 is switched to PW2 as an example.
  • the MAC address forwarding table needs to be switched.
  • the PE3 needs to perform the migration of the MAC address forwarding table. For details, see Figure 6.
  • Step 201 The PE1 performs the MAC address forwarding table update manner according to the method for switching the MAC address forwarding table provided in Embodiment 1, and is not described in detail herein;
  • Step 202 After detecting that PW1 fails, PE3 switches from PW1 to PW2, that is, PE3 stops using PW1, and advertises and activates PE4.
  • Step 203 The PE3 sends a MAC address forwarding table migration message to the PE4, where the message includes the MAC address list, the identifiers of the primary PW and the standby PW (PW1 and PW2 in this embodiment), and the MAC address list is the port in the MAC address forwarding table of the PE3. Partial or all entries corresponding to PW1;
  • the PE3 first checks its own MAC address forwarding table, and finds the entry corresponding to the PW1 in the MAC address forwarding table, and sends the entry to the PE4.
  • the PE3 in this embodiment sends the entry to the PE4.
  • Forwarding table migration message implementation the message can be expressed in the form of MAC_Transfer (PWl, PW2, Mac_List), where PWl and PW2 are the primary PW and the alternate PW identifier respectively, and the Mac_List is the MAC list, that is, the corresponding port in the MAC address forwarding table is PW1.
  • MAC address if MAC_List is empty, it means that all ports in the MAC address forwarding table are PW1 addresses.
  • the forwarding table migration message can be encapsulated by the LDP protocol message format and communicated through the LDP session.
  • PE3 may only send the complete MAC address forwarding table to its standby device PE4 for the first time, and save it by PE4, and then only periodically it.
  • the updated part of the MAC address forwarding table is sent to the PE4; or the MAC address forwarding table is completely sent to the PE4 only when the AC/PW fails and the management platform switches.
  • Step 204 After receiving the MAC address forwarding table migration message sent by the PE3, the PE4 adds the MAC address list in the message to its own MAC forwarding table, and sets the port of the MAC address to PW2.
  • the MAC address forwarding table of PE4 itself may be an empty table or an empty table before the migration. Whether the MAC address forwarding table is an empty table or a non-empty table, the updated entry can be added to the MAC address forwarding table. Save it.
  • the PE1 in this embodiment may also send a forwarding table update notification to the PE3 after detecting that the PW1 is faulty.
  • the notification may be implemented by using a MAC_Transfer (PW1, PW2, MAC_List) message. .
  • PE1 and PE2 are respectively a pair of primary PEs.
  • the device and the standby PE device, PE3 and PE4 are a pair of primary PE devices and standby PE devices respectively.
  • PW1 and PW2 are the primary PW and the standby PW respectively.
  • AC1 in Figure 7 fails, and PW1 switches to PW2.
  • the switch from PW1 to PW2 causes the local PE device to change. Therefore, both PE1 and PE3 perform the forwarding table migration operation performed by PE3 to implement the MAC address of the standby PE device. The update of the address forwarding table is not detailed here.
  • the PE3 sends only the MAC address list corresponding to the PW1, and the identifier information of the PW1 and the PW2 to the PE4.
  • the PE4 adds the received MAC address list to its own MAC address forwarding table, and then sets its port to PW2.
  • the PE3 directly replaces the PW1 in the port entry with the PW2, and then sends the replaced MAC address forwarding table to the PE4.
  • the PE4 in the step 204 receives the entry sent by the PE1. , directly save the MAC address forwarding table; the other steps are unchanged.
  • PE3 directly sends its own MAC address forwarding table to PE4, and indicates the identification information of the primary PW and the standby PW, that is, PW1 and PW2.
  • PE4 After receiving the MAC address forwarding table, PE4 replaces the entry with the port PW1 with The port is PW2.
  • the PE device after switching from PW1 to PW2, the PE device updates its MAC address forwarding table through the migration or switching operation of the MAC address forwarding table, and the PE device forwards the packet through the PW2, thereby avoiding the recovery and re-learning of the MAC address.
  • the process and its resulting packet flooding reduce the impact on network services, improve the update efficiency of the MAC address forwarding table, and save the bandwidth of the core network.
  • This embodiment provides a method for updating a forwarding table in an Ethernet service transmission. See the network structure shown in Figure 8.
  • the functions in PE1, PE2, PW1, and PW2 are the same as those in Figure 7.
  • the PE1 in the 8 is faulty, and the PW1 is switched to the PW2 as an example.
  • the method for updating the MAC address forwarding table includes:
  • PE2 and PE3 detect that lj PE1 fails, switch PW1 to lj PW2, S ⁇ PE2 activates PW2, and PE3 performs the forwarding operation of the forwarding table.
  • PE3 will update the MAC address forwarding table on PE4 according to the migration operation of the MAC address forwarding table provided in Embodiment 2.
  • the PE2 can forward the packet to the PE2.
  • the MAC address forwarding table is configured to forward the MAC address forwarding table corresponding to the PW1 to the PE2. And get it.
  • the PE1 fails to obtain the MAC address forwarding table from PE1, that is, the MAC address forwarding table of PE2 is empty, you need to obtain the MAC address forwarding table through self-learning.
  • PE2 directly enables its own MAC address forwarding.
  • the PE3 updates the MAC address forwarding table.
  • the PE2 and the PE4 forward the packet through the PW2. This prevents the MAC address recovery and re-learning process and the packet flooding caused by the MAC address.
  • the impact of the network service improves the update efficiency of the MAC address forwarding table and saves the bandwidth of the core network.
  • This embodiment provides a method for updating a forwarding table in an Ethernet service transmission.
  • This embodiment uses a specified device (for example, a management platform) to perform PW handover determination and determination as an example.
  • the management platform may Perform manual PW switching operations without failure. You can also perform PW switching responses after collecting fault messages (for example, when AC PW or PE devices fail).
  • the designated device is responsible for notifying the corresponding PE device to perform MAC address forwarding table migration or MAC address forwarding table switching operation, and the corresponding PE device performs MAC address forwarding table migration or MAC address forwarding table switching operation according to the notification of the specified device.
  • the specific MAC address forwarding table migration or MAC address forwarding table switching operation is similar to Embodiment 1 or 2 above.
  • the networking diagram of Figure 9 is used as an example to describe the migration and switching process of the MAC address forwarding table.
  • the PE1 PE2 is the primary PE and the standby PE.
  • the PW1 is the PW of the PW3.
  • PW, before the PW switchover, the MAC address forwarding table on PE1 is shown in Table 1.
  • the MAC address forwarding table on PE3 is shown in Table 2.
  • the MAC address forwarding table on PE4 is shown in Table 3. (For ease of understanding, Added source MAC address column):
  • the specific device for example, the network management device
  • the PW switch causes the local PE device to change.
  • the PE1 changes the local PE device.
  • PE1 becomes PE2.
  • the local PE device does not change. Therefore, the MAC address forwarding table migration notification is sent to PE1, and the MAC address forwarding table switching notification is sent to PE3 and PE4.
  • the migration method of the MAC address forwarding table is sent to PE1, and the MAC address forwarding table switching notification is sent to PE3 and PE4.
  • PE1 After PE1 receives the MAC address forwarding table migration notification sent by the specified device, PE1 sends a MAC address forwarding table migration message to PE2, Mac_Transfer (PWl, PW3 : MAC Listl), Mac_Transfer (PW2, PW4, MAC_List2), where PW1 PW2 and P PW3 PW4 are the identifiers of the primary PW and the standby PW respectively, and MAC_List1 is the partial or all entries corresponding to the PW1 in the MAC address forwarding table of the PE1, and the MAC_List2 is the MAC address forwarding table of the PE1.
  • the intermediate port is a partial or all entry corresponding to PW2.
  • the two MAC migration messages can also be encapsulated in the same LDP message. give away;
  • PE2 After receiving the MAC address forwarding table migration message sent by PE1, PE2 adds the MAC address list to its own MAC address forwarding table, and replaces the port corresponding to the MAC address with PW1 and PW3 and PW2.
  • the PE3 forwards the PW1 in the MAC address table of the MAC address. Replace with PW3 PE4 and replace PW2 in its own MAC address forwarding table with PW4.
  • the MAC address forwarding table on PE2 is shown in Table 4.
  • the MAC address forwarding table on PE3 is shown in Table 5.
  • the MAC address forwarding table on PE4 is shown in Table 6. Understand, the source address column has been added):
  • the process of updating the MAC address forwarding table in the foregoing PE1 to PE2 is the migration process of the MAC address forwarding table.
  • the process of updating the MAC address forwarding table of the PE3 and the PE4 is the process of switching the MAC address forwarding table, and the MAC address forwarding table migration and handover process.
  • the specific implementation is similar to that of Embodiment 1 and Embodiment 2, and will not be described in detail herein.
  • the designated device notifies the corresponding PE device to perform the MAC address forwarding table migration or handover process, thereby implementing the MAC address forwarding table update, avoiding the MAC address recovery and re-learning process and the packet flooding caused by the MAC address reduction.
  • the impact on the network service improves the update efficiency of the MAC address forwarding table and saves the bandwidth of the core network.
  • multiple ACs may access the same PE device and transmit data packets through the same PW.
  • Ethernet services from the same AC may be used.
  • it may only be necessary to switch or migrate a partial MAC address table corresponding to a specific AC which may be implemented by sending the partial MAC address list, which is similar to the MAC reclaim message defined in RFC 4762 in the prior art.
  • a list of MAC addresses in If you need to switch all the MAC forwarding tables of the corresponding PW, send an empty MAC list.
  • the forwarding table update notification can use the MAC_Switch (PWl, PW2, Mac List) message.
  • PE1 PE2 is the primary PE device and the standby PE device respectively.
  • PW1 PW2 is the primary PW and the standby PW Mac_List are MAC lists.
  • the PE1 in Figure 7 is used as an example. If PE1 is connected to multiple CEs, some CEs need to be protected, such as CE1 in Figure 7. Some CEs do not need to be protected. In this case, PE1 forwards the MAC address from its own MAC address. After the port corresponding to the PW1 entry is found, the MAC address entry corresponding to the MAC address of the CE device that is protected on the PE2 is sent to the PE2.
  • the above MAC address forwarding table migration and switching can be divided into active mode and passive mode.
  • active mode the PE device actively performs MAC address forwarding table migration and switching. This device is usually the initiator in the PW redundant switch independent mode or the primary PE device in the master-slave mode.
  • passive mode the PE device is in the passive mode. After receiving the forwarding table migration or switching notification message, the forwarding table is migrated or switched. After receiving the notification, the PE device replaces the primary PW in the forwarding table with the standby PW.
  • this embodiment provides an operator edge device, including:
  • the fault processing module 301 is configured to: when detecting that the link or the PE device is faulty, switch the primary link to the standby link; and the forwarding table update module 302 is configured to: when the fault processing module 301 switches from the primary link to the standby link, Will be its own
  • the port corresponding to the primary link in the MAC address forwarding table is updated to the port corresponding to the standby link.
  • the port in this embodiment can be either a physical port or a logical port.
  • the link in the embodiment of the present invention may be a physical link or a logical link, where there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port), for example, for the chain.
  • the technical solution provided by this embodiment still applies to the case where the physical path is a physical link and the port is a physical port.
  • the carrier edge device in the embodiment first checks its own MAC address forwarding table, and finds each corresponding entry of the port as the primary pseudowire from the MAC address forwarding table, and then each The main pseudowire in the entry is replaced with an alternate pseudowire. It will not be detailed here.
  • the port corresponding to the primary link in the MAC address forwarding table of the device is updated to the port corresponding to the standby link;
  • the packet is forwarded, which avoids the recovery and re-learning process of the MAC address and the packet flooding caused by the MAC address, reduces the impact on the network service, improves the update efficiency of the MAC address forwarding table, and saves the bandwidth of the core network.
  • this embodiment provides an operator edge device, including:
  • the determining module 401 is configured to determine, when the primary link is switched to the standby link, whether the corresponding PE device on the standby link is itself;
  • the processing module 402 is configured to: when the judgment result of the determining module 401 is that the corresponding PE device on the standby link is the self, replace the port corresponding to the primary link in the MAC address forwarding table of the user with the backup link. Port; otherwise, send a MAC address forwarding table migration message to the corresponding PE device on the standby link; the message may include a MAC address The identification of the list, primary link, and alternate link.
  • the port in this embodiment can be either a physical port or a logical port.
  • the link in this embodiment may be a physical link or a logical link, and there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port), for example, for the chain.
  • the technical solution provided by this embodiment still applies to the case where the physical path is a physical link and the port is a physical port.
  • the update method used by the PE3 in the embodiment 1, 2, 3 or 4 may be adopted, that is, when the local PE device on the standby pseudowire is the self.
  • Check the MAC address forwarding table of the device find the corresponding entry of the port as the primary pseudowire from the MAC address forwarding table, and then replace the primary pseudowire in each entry with the alternate pseudowire;
  • the local PE device is not the local device, the migration of the MAC address forwarding table is initiated.
  • the operator edge device in the embodiment determines to perform the switching or migrating operation of the MAC address forwarding table according to the judgment result of the determining module 401, thereby avoiding the process of recovering and re-learning the MAC address and the packet flooding caused by the MAC address, and reducing the network to the network.
  • the impact of the service improves the update efficiency of the MAC address forwarding table and saves the bandwidth of the core network.
  • this embodiment provides an operator edge device, including:
  • the receiving module 501 is configured to receive a MAC address forwarding table migration message sent by the corresponding PE device on the primary link.
  • the processing module 502 is configured to: after receiving the MAC address forwarding table migration message, the receiving module 501, the MAC address forwarding table migration message is sent.
  • the MAC address included in the MAC address is added to its own MAC address forwarding table, and the ports of these MAC addresses are set to the ports corresponding to the alternate link.
  • the port in this embodiment can be either a physical port or a logical port.
  • the link in this embodiment may be a physical link or a logical link, and there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port), for example, for the chain.
  • the technical solution provided by this embodiment still applies to the case where the physical path is a physical link and the port is a physical port.
  • the operator edge device in this embodiment is a standby carrier edge device.
  • the MAC address forwarding table is updated according to the MAC address forwarding table migration message, thereby avoiding the MAC address learning process and the grouping caused by the MAC address forwarding table.
  • Flooding reduces the impact on network services, improves the update efficiency of the MAC address forwarding table, and saves the bandwidth of the core network.
  • the embodiment provides an update system for forwarding tables in an Ethernet service transmission, including: a primary carrier edge device 601, configured to determine standby operations on the standby link when the primary link is switched to the standby link. Whether the quotient edge device 602 is itself; if yes, replace the port corresponding to the primary link in the own MAC address forwarding table with the port corresponding to the standby link; otherwise, send the MAC address to the alternate carrier edge device 602. Post a migration message;
  • the standby carrier edge device 602 is configured to: after receiving the MAC address forwarding table migration message of the primary carrier edge device 601, add the MAC address included in the MAC address forwarding table migration message to its own MAC address forwarding table, and add these MAC addresses.
  • the port of the address is set to the port corresponding to the alternate link.
  • the port in this embodiment can be either a physical port or a logical port.
  • the link in this embodiment may be a physical link or a logical link, and there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port), for example, for the chain.
  • the technical solution provided by this embodiment still applies to the case where the physical path is a physical link and the port is a physical port.
  • the operation of the MAC address forwarding table update performed by the primary carrier edge device 601 and the standby carrier edge device 602 is similar to the operation of updating the PE1 and the PE2 in the embodiment 2, and details are not described herein.
  • the system in this embodiment updates the MAC address forwarding table by using the primary carrier edge device 601 and the standby carrier edge device 602, and updates the primary PW to the standby PW, thereby avoiding the process of recovering and re-learning the MAC address.
  • the resulting packet flooding reduces the impact on the network service, improves the update efficiency of the MAC address forwarding table, and saves the bandwidth of the core network.
  • this embodiment provides an update system for forwarding table in Ethernet service transmission, including: determining device 701, when determining that the primary link is switched to the standby link, determining whether the primary link is switched to the standby link.
  • the corresponding PE device changes, if yes, sends a MAC address forwarding table migration notification to the primary carrier edge device 702 on the primary link; otherwise, sends a MAC address forwarding table switch to the primary carrier edge device 702 on the primary link.
  • the primary carrier edge device 702 is configured to send a MAC address forwarding table migration message to the alternate carrier edge device 703 on the standby link after receiving the MAC address forwarding table migration notification sent by the determining device 701; After the MAC address forwarding table switching notification sent by the device 701, the port corresponding to the primary link in the MAC address forwarding table of the device 701 is replaced with the port corresponding to the standby link;
  • the backup carrier edge device 703 is configured to receive the MAC address forwarding table migration of the primary carrier edge device 702. After the information is added, the MAC address included in the MAC address forwarding table migration message is added to its MAC address forwarding table, and the ports of these MAC addresses are set to the ports corresponding to the standby link.
  • the port in this embodiment can be either a physical port or a logical port.
  • the link in this embodiment may be a physical link or a logical link, and there are multiple logical links, including a PW.
  • the primary and backup links are pseudowires
  • the port is a logical port (that is, a pseudowire, because the pseudowire itself can be either a logical link or a logical port), for example, for the chain.
  • the technical solution provided by this embodiment still applies to the case where the physical path is a physical link and the port is a physical port.
  • the system in this embodiment triggers the primary carrier edge device 702 to perform the migration operation or the switching operation of the MAC address forwarding table by the determining device 701, thereby avoiding the process of recovering and re-learning the MAC address and the packet flooding caused by the MAC address, and reducing the pairing
  • the impact of the network service improves the update efficiency of the MAC address forwarding table and saves the bandwidth of the core network.
  • a full-grid connection is generally used between PEs, that is, there is only one pseudowire between each pair of PEs.
  • all the message advertisements in the above embodiments may also carry the identification information of the primary node and the standby node and the MAC address list, and the received operator edge device may obtain the corresponding information according to the identification information of the primary node and the standby node.
  • the primary link and the alternate link, and the MAC address update is performed according to the method of the above embodiment.
  • the information of the link or the node is sufficient to determine the information of the link or the node according to the channel on which the message is located, the message only needs to carry information of one node or one link under optimization.
  • the technical solution provided by the foregoing embodiment can be applied to the networking situation with a redundant architecture, for example, the networking structure in the case of active/standby protection or load balancing, and the redundancy protection relationship between the PE device and the PW group can avoid the Ethernet.
  • the MAC address recovery and self-learning process under the multi-point service transmission in order to reduce the communication delay, avoid excessive bandwidth consumption caused by the MAC address self-learning process, and increase the reliability and scalability of the Ethernet multi-point service.
  • All or part of the technical solutions provided by the above embodiments may be implemented by software programming, and the software program is stored in a readable storage medium such as a hard disk, an optical disk or a floppy disk in a computer.

Description

以太网业务传送中转发表的更新方法、 设备和系统 本申请要求于 2008年 5月 16日提交中国专利局、 申请号为 200810081885.7、 发明名 称为 "以太网业务传送中转发表的更新方法、 设备和系统" 的中国专利申请的优先权, 其 全部内容通过引用结合在本申请中。
技术领域
本发明涉及通信领域, 特别涉及一种以太网业务传送中转发表的更新方法、 设备和系 统。
背景技术
以太网是目前部署最为广泛的局域网技术, 但是它存在一些限制, 例如 VLAN(Virtual Local Area Network,虚拟局域网)地址空间有限、生成树协议扩展性有限、 MAC(Media Access Control, 媒体访问控制) 地址不易汇聚等缺陷。
为了突破以太网交换的限制, 人们开发了若干能够提供以太网多点连接服务的新兴技 术,一种是 L2VPN (Layer 2 Virtual Private Network,二层虚拟专用网络)中的 VPLS (Virtual Private LAN Service , 虚拟专用 LAN 服务) 技术, 它可以通过供应商的 IP/MPLS (Multi-protocol Label Switching, 多协议标签互换协议)网络支持的仿真以太网广播域将多 个站点连接在一起。 另一种是基于以太网的 PBBN (Provider Backbone Bridge Network, 运 营商骨干网桥网络)技术, 它通过两层 MAC地址封装, 为以太网客户提供城域汇聚和业务 传送的功能。
如图 1所示,为 VPLS的网络架构参考模型,其中,客户的以太网业务通过 CE(Customer Edge, 客户边缘) 设备接入到 PE (Provider Edge, 运营商边缘) 设备, PE设备之间经过一 个或多个 P设备 (例如, P路由器) 连接在一起。 一个 VPLS实例由多个 PE设备组成, 当 PE设备通过 AC (Attach Circuit, 接入电路) 接收到 CE设备发来的以太网帧后, VPLS转 发器(具有 MAC地址转发表的设备或模块)将根据以太网帧中的目的 MAC地址等信息获 取在 MPLS网络上进行传送的 PW (Pseudowire, 伪线), 然后通过 PW复用到分组交换网 络的隧道中完成二层报文的透传。
对于 PBBN网络, 可以采用类似的架构来完成以太网多点业务的传送, 只是其中的伪 线属于另一种逻辑通道, 而复用的网络隧道则是以太网交换路径。
上述 MAC地址转发表是通过一种 MAC地址自学习机制建立的, 类似于以太网的自学 习过程, 简单介绍如下:
当网络边缘设备 (PE) 从一条 AC/PW (为方便起见, 将 AC或 PW到 PE的连接点都 称为端口, 它可能是一个物理端口, 也可能只是一个逻辑端口, 例如 PW所对应的逻辑端 口)接收到分组时, 如果该分组的目的 MAC地址为广播地址或是未知地址, 则该分组将洪 泛到所有其它端口; 如果该分组的源 MAC地址未知 (指在 MAC地址转发表中找不到对应 于它的表项), 则将该源 MAC地址与该分组的接收端口绑定在一起, 保存在 MAC地址转 发表中, 以后目的 MAC地址为该地址的新分组就发送到该端口。 该过程也称为 MAC地址 自学习过程。
当网络拓扑发生变化(例如 AC或 PW发生切换) 时, 上述 MAC地址与端口的绑定关 系可能变成无效。 如图 2所示, CE1设备本来通过 AC1接入 PE1设备, 业务报文由 PW1 传送到 PE3设备, 最后到达客户 CE2设备上。 当 AC1发生故障时, CE1切换到 AC2来接 入 PE2设备, 业务由 PW2传送到 PE3设备, 最后到达客户 CE2设备上。 在此过程中, 与 上述切换操作相关的 PE1设备、 PE2设备和 PE3设备上的 MAC地址转发表都需要进行更 新,以便反映新的拓扑关系,现有技术中提供了以下几种方式进行 MAC地址转发表的更新:
1 ) 地址老化机制
在地址老化机制中, 若 MAC地址转发表中的 MAC地址在一定时间内未使用, 则删除 该 MAC地址对应的表项, 再重新通过自学习过程来获取新的 MAC地址转发表项, 进而实 现 MAC地址转发表的更新。
2) MAC地址回收机制
当网络拓扑发生变化时, PE设备向 VPLS实例中的其它所有对端 PE设备发送一条 MAC 地址回收消息, 该消息包含欲清除的 MAC地址列表, 其它 PE设备根据该消息删除相应的 MAC地址表项。 当 PE设备发送的是一条携带空 MAC地址表的地址回收消息时, 则清除 所有与该 VPLS实例相联的 MAC地址, 除了从接收该消息的 PW上学习到的 MAC地址。
3 ) 特定 MAC地址清除机制
通过扩展 VPLS采用的 LDP (Label Distribution Protocol, 标签分发协议) 信令, 在信 令中增加一种携带 PE设备标识符的 TIN (Type-Length- Value, 类型 -长度 -值) 字段, 该字 段采用通用的协议字段表示形式,可以指定只清除从特定 PE设备学习到的所有 MAC地址, 以便大幅缩小 MAC地址清除的范围。
在实现本发明的过程中, 发明人发现现有技术至少存在以下问题: 上述方法都需要清除 MAC地址转发表中的对应表项,再通过自学习的方式重新学习获 得新的 MAC地址转发表。 由于自学习过程中将进行分组洪泛, 将在核心网络产生大量的冗 余报文, 增加了网络带宽的消耗。 另外在老化机制中, MAC地址转发表项的更新延时大, 可能导致较长时间内的分组未能发送到新拓扑的正确 PE设备上, 而是发送到旧拓扑的 PE 设备上。 发明内容
为了降低 MAC地址转发表更新过程的延时和对网络带宽的消耗,本发明实施例提供了 一种以太网业务传送中转发表的更新方法、 设备和系统。 所述技术方案如下:
一种以太网业务传送中转发表的更新方法, 所述方法包括:
当主链路切换到备用链路时, 运营商边缘 (PE) 设备根据所述主链路、 所述备用链路 以及媒体访问控制 (MAC) 地址列表更新自身的 MAC地址转发表。
一种运营商边缘设备, 所述设备包括:
故障处理模块 (301 ), 用于链路或运营商边缘 (PE) 设备出现故障时, 将主链路切换 到备用链路;
转发表更新模块 (302), 用于所述故障处理模块 (301 ) 从主链路切换到备用链路时, 根据所述主链路、 所述备用链路以及媒体访问控制 (MAC) 地址列表更新自身的 MAC地 址转发表。
一种运营商边缘设备, 所述设备包括:
判断模块 (401 ), 用于主链路切换到备用链路时, 判断所述备用链路上是否有一端运 营商边缘 (PE) 设备为自身;
处理模块 (402), 用于当所述判断模块 (401 ) 的判断结果是所述备用链路有一端 PE 设备为自身时, 将自身的媒体访问控制 (MAC) 地址转发表中的与所述主链路对应的表项 的端口替换为与所述备用链路对应的端口; 否则, 向所述备用链路上本地对应的 PE设备发 送 MAC地址转发表迁移消息。
一种运营商边缘设备, 所述设备包括:
接收模块 (501 ), 用于接收包含媒体访问控制 (MAC) 地址列表的消息;
处理模块 (502), 用于所述接收模块 (501 ) 收到所述消息后, 根据所述 MAC地址列 表中包含的 MAC地址更新自身的 MAC地址转发表, 并将所述 MAC地址的端口设置为与 所述备用链路对应的端口。
一种以太网业务传送中转发表的更新系统, 所述系统包括: 主运营商边缘设备 (601 ), 用于主链路切换到备用链路时, 判断所述备用链路是否有 一端运营商边缘设备 (PE) 为自身; 如果是, 将自身的媒体访问控制 (MAC) 地址转发表 中的与所述主链路对应表项的端口替换为与所述备用链路对应的端口; 否则, 向备用运营 商边缘设备发送 MAC地址转发表迁移消息;
备用运营商边缘设备 (602), 用于收到所述主运营商边缘设备 (601 ) 的 MAC地址转 发表迁移消息后, 将所述 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC 地址转发表中, 并将所述 MAC地址表项的端口设置为与所述备用链路对应的端口。
一种以太网业务传送中转发表的更新系统, 所述系统包括:
判断装置 (701 ), 用于主链路切换到备用链路时, 判断所述主链路切换到所述备用链 路是否导致运营商边缘 (PE) 设备发生变化, 如果是, 向所述主链路上的主运营商边缘设 备 (702)发送媒体访问控制 (MAC)地址转发表迁移通知; 否则, 向所述主链路上的主运 营商边缘设备 (702) 发送 MAC地址转发表切换通知;
主运营商边缘设备(702),用于收到所述判断装置发送的 MAC地址转发表迁移通知后, 向所述备用链路上的对应备用运营商边缘设备 (703 ) 发送 MAC地址转发表迁移消息; 以 及收到所述判断装置 (701 ) 发送的 MAC地址转发表切换通知后, 将自身的 MAC地址转 发表中的与所述主链路对应表项的端口替换为与所述备用链路对应的端口;
备用运营商边缘设备 (703 ), 用于收到所述主运营商边缘设备 (702) 的 MAC地址转 发表迁移消息后, 将所述 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC 地址转发表中, 并将所述 MAC地址的表项端口设置为与所述备用链路对应的端口。
本发明实施例提供的技术方案的有益效果是:
通过主链路切换到备用链路时, 备用链路上的 PE设备将自身的 MAC地址转发表中的 与主链路对应的端口更新为与备用链路对应的端口,避免了 MAC地址的回收和重学习过程 及其引起的分组洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节 省了核心网络的带宽。 附图说明
图 1是现有技术提供的 VPLS的网络架构参考模型示意图;
图 2是现有技术提供的组网结构示意图;
图 3是本发明实施例 1提供的组网结构示意图;
图 4是本发明实施例 1提供的以太网业务传送中转发表的更新方法流程图;
图 5是本发明实施例 2提供的组网结构示意图; 图 6是本发明实施例 2提供的以太网业务传送中转发表的更新方法流程图;
图 7是本发明实施例 2提供的另一种组网结构示意图;
图 8是本发明实施例 3提供的组网结构示意图;
图 9是本发明实施例 4提供的组网结构示意图;
图 10是本发明实施例 4提供的图 9进行 PW切换后的组网结构示意图;
图 11是本发明实施例 5提供的运营商边缘设备结构示意图;
图 12是本发明实施例 6提供的运营商边缘设备结构示意图;
图 13是本发明实施例 7提供的运营商边缘设备结构示意图;
图 14是本发明实施例 8提供的以太网业务传送中转发表的更新系统结构示意图; 图 15是本发明实施例 9提供的以太网业务传送中转发表的更新系统结构示意图。 具体实施方式
为使本发明的目的、 技术方案和优点更加清楚, 下面将结合附图对本发明实施方式作 进一步地详细描述。
本发明实施例通过主链路切换到备用链路时,与备用链路相连的 PE设备将自身的 MAC 地址转发表中的与主链路对应的端口更新为与备用链路对应的端口, 实现了在主链路切换 到备用链路后, 不需要再进行 MAC地址回收与自学习的过程, 降低 MAC地址转发表更新 过程的延时和对网络带宽的消耗。
本发明实施例提供的 MAC地址转发表中包括: 目的 MAC地址和端口信息, 其中, 端 口既可以是物理端口, 也可以是逻辑端口。
本发明实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包括 PW等。 本发明以下实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪 线本身既可以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理 链路、 端口为物理端口的情况, 本发明实施例提供的技术方案仍然适用。 本发明以下实施 例中的逻辑端口用 PW表示, 进行报文转发时, 根据报文的目的 MAC地址在 MAC地址转 发表中查找对应的表项, 该表项中的端口对应的 PW即为传送该报文所使用的 PW。
为了保证以太网业务的高可用性, VPLS的部署一般采用冗余网络架构, 当部分接入电 路 AC、 PW以及 PE设备发生故障时, 可以继续使用备用 AC、 PW或 PE设备。 另外, 多 个冗余 PW还可以形成负载分担的集合, 使得以太网业务可以在这些 PW上同时传送。 这 种互为冗余备份或负载分担关系的 PE设备和 PW各自可组成保护组,从而一个保护组内的 所有 PE设备或 PW构成冗余备份或负载分担的关系。群组关系的建立可采用管理配置或者 通过控制信令方式来实现。
例如, 如果采用管理配置, 可以通过管理平台来进行人工设定, 也可以通过 PCE (Path Computation Element, 路径计算单元)等专门路由平台来计算得到。 当建立 PW时, 同时为 该 PW配置一个备用 PW, 并将主 PW和备用 PW关联在同一个保护组中, 配置之后, 在 PE设备上保存有各个 VPLS实例相关的 PW保护组及其两端 PE设备识别符的绑定关系, 即由 PW可以获知对应的 PE设备。 这样, 可以保证数据业务在主 PW故障时, 仍能通过备 用 PW转发。
通常所有 PE设备之间会建立全连接的 LDP会话, SP, 每两个 PE设备之间都有一个 LDP会话, 以便通过这些 LDP会话来进行信令控制。
通过网络管理平台或人工指令, 或者在发生下列故障的情况下, 主 PW会切换到备用
PW:
主 AC 发生故障, 其中, AC 的故障检测可以通过物理信号检测、 APS ( Automatic Protection Switching, 自动保护倒换) 协议、 LACP (Link Aggregation Control Protocol, 链 路聚合控制协议) 协议、 802.1ag CC (Continuity and Connectivity Check, 连续性检测) 报 文等方式实现;
主 PE设备或主 PW发生故障, 其中, PE设备或 PW的故障检测可以通过路由协议、 VCCV ( Virtual Circuit Connection Verification, 虚电路连接验证) 禾 P BFD (Bidirectional Forwarding Detection, 双向转发检测) 等机制实现。
当 AC、 PW或 PE设备发生故障时, 则业务的传送将切换到新的 AC、 PW或 PE设备 上,以保证业务尽量不受影响。例如有的双归属保护协议能够进行 PE设备之间的状态同步, 保证当主 AC或主 PE设备发生故障时, 自动让 CE通过备用 AC、 备用 PW或备用 PE设备 传送信流。 其中, 双归属保护协议目前通常为私有协议。
当 PW进行切换时, 对于主 PW的本地 PE设备来说, 根据备用 PW是否经过该 PE设 备, 即切换前后是否经过同一个 PE设备, PW切换带来的 MAC地址转发表的更新过程可 以分为 MAC地址转发表迁移(本地 PE设备改变)和 MAC地址转发表切换(本地 PE设备 不变) 两种情形, 具体内容在下面的实施例中详细进行说明。 实施例 1
本实施例提供了一种以太网业务传送中转发表的更新方法, 参见图 3 提供的组网结构 示意图, 其中, PW1、 PW2分别是主 PW和备用 PW, 本实施例以图 3中的 PW1发生故障, 业务由 PW1切换到 PW2为例进行说明, 因为从 PW1切换到 PW2, 未导致本地 PE设备发 生变化, 本地 PE设备切换前后都是 PE1, 所以本实施例采用 MAC地址转发表切换方法进 行 MAC地址转发表的更新, 参见图 4, MAC地址转发表的更新方法包括:
步骤 101 : PE1和 PE3检测到 PW1发生故障后, 从 PW1切换到 PW2, 即停止对 PW1 的使用, 激活 PW2;
步骤 102: PE1和 PE3判断出 PW切换前后本地 PE设备没有发生改变, 分别将自身的
MAC地址转发表中的 PW1更新为 PW2。
在 LDP信令中, 通用 PW标识符会包含其端点 (PE) 的标识 (例如地址), 因此只要 将所检测到的 PW1的端点标识 (地址) 与 PE节点自身地址进行比较即可, 如果相同, 则 说明 PW切换前后本地 PE设备没有发生改变, 不同, 则说明已改变。
PE1和 PE3首先检查自身的 MAC地址转发表,从 MAC地址转发表中找出端口为 PW1 的每个对应表项, 然后将各表项中的 PW1替换为 PW2。
完成上述 MAC地址转发表的更新后, PE1 和 PE3将通过 PW2传送目的地址为上述 MAC地址的数据报文。
或者, 如果 PE3不主动进行上述检测和 MAC地址转发表的更新时, PE1更新完自身的 MAC地址转发表后, 可以向 PE3发送转发表更新通知;
其中, 转发表更新通知可以采用 MAC_Switch (PW1 , PW2, MAC_List) 消息形式实 现, 表示将 MAC地址转发表中对应于 MAC_List中地址的端口从 PW1更新为 PW2, 该消 息可以通过 LDP中的消息格式封装, 并通过 LDP会话进行通信, 若 MAC_List为空, 则表 示 MAC地址转发表中所有端口为 PW1的地址;
PE3收到 PE1发送的转发表更新通知后,根据通知中的内容(例如, MAC_Switch(PWl,
PW2, MAC_List)) 将自身 MAC地址转发表中对应于 MAC_List中地址的每个表项的端口 替换为 PW2, 若 MAC_List为空, 则将 MAC地址转发表中所有端口为 PW1的每个表项的 端口替换为 PW2。
本实施例在从 PW1切换到 PW2后, PE1和 PE3通过修改自身的 MAC地址转发表,进 而 PE1和 PE3通过 PW2转发报文, 避免了 MAC地址的回收和重学习过程及其引起的分组 洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的 带宽。 实施例 2
本实施例提供了一种以太网业务传送中转发表的更新方法, 参见图 5 提供的组网结构 示意图, 其中, PE3、 PE4分别是主 PE设备和备用 PE设备, PW1、 PW2分别是主 PW和 备用 PW, PEl通过 PWl和 PW2分别与 PE3和 PE4相连, 本实施例以图 5中的 PW1发生 故障, 由 PW1切换到 PW2为例进行说明, 根据 PW切换前后是否导致本地 PE设备发生变 化, PE1需要进行 MAC地址转发表的切换, PE3需要进行 MAC地址转发表的迁移, 参见 图 6, 下面简单描述一下该组网场景下的 MAC地址转发表的更新方法:
步骤 201: PE1按照实施例 1提供的 MAC地址转发表的切换方法进行 MAC地址转发 表的更新, 这里不再详述;
步骤 202: PE3检测到 PW1发生故障后, 从 PW1切换到 PW2, 即 PE3停止对 PW1的 使用, 通告并激活 PE4;
步骤 203 : PE3向 PE4发送 MAC地址转发表迁移消息, 该消息包含 MAC地址列表、 主 PW和备用 PW的标识(本实施例为 PW1和 PW2), MAC地址列表为 PE3的 MAC地址 转发表中端口为 PW1对应的部分表项或全部表项;
PE3首先检查自身的 MAC地址转发表, 从 MAC地址转发表中找出端口为 PW1对应 的表项, 将这些表项发送给 PE4; 本实施例中的 PE3将表项发送给 PE4时, 可以采用转发 表迁移消息实现,该消息可以表示为 MAC_Transfer (PWl, PW2, Mac_List)形式,其中 PWl 和 PW2分别为主 PW和备用 PW标识, Mac_List为 MAC列表, 即 MAC地址转发表中端 口为 PW1的相应 MAC地址, 若 MAC_List为空, 则表示 MAC地址转发表中所有端口为 PW1 的地址。 该转发表迁移消息可通过 LDP协议消息格式封装, 并通过 LDP会话进行通 信。
因为 MAC地址转发表可能比较大, 并且可能会随着时间发生改变, PE3可以仅在首次 将完整的 MAC地址转发表发送给其备用设备 PE4, 由 PE4进行保存, 随后每隔一段时间仅 将其 MAC地址转发表中更新的部分发送给 PE4; 或者只在 AC/PW发生故障、 管理平台进 行切换时, 将 MAC地址转发表完整地发送到 PE4上。
步骤 204: PE4收到 PE3发送的 MAC地址转发表迁移消息后, 将消息中的 MAC地址 列表增加到自身的 MAC转发表中, 并将这些 MAC地址的端口设置为 PW2;
PE4自身的 MAC地址转发表在迁移之前可以是一张空表,也可以不是空表,无论 MAC 地址转发表是空表还是非空表,都可以将更新后的表项添加在其 MAC地址转发表中进行保 存。
同实施例 1类似,本实施例中的 PE1也可以在检测到 PW1发生故障后, 向 PE3发送转 发表更新通知, 该通知可以采用 MAC_Transfer (PWl, PW2, MAC_List) 消息实现, 这里 不再详述。
上述方法还可以应用于图 7提供的组网结构示意图, 其中, PE1、 PE2分别是一对主 PE 设备和备用 PE设备, PE3、 PE4分别是一对主 PE设备和备用 PE设备, PW1、 PW2分别是 主 PW和备用 PW, 本实施例以图 7中的 AC1发生故障, 由 PW1切换到 PW2为例进行说 明, 对于 PE1和 PE3来说, 因为从 PW1切换到 PW2, 都导致了本地 PE设备发生变化, 所 以 PE1和 PE3都将执行上述 PE3进行的转发表迁移操作,实现其备用 PE设备的 MAC地址 转发表的更新, 这里不再详述。
上述步骤 203中 PE3只将 PW1对应的 MAC列表、 PW1和 PW2的标识信息发送给 PE4, PE4将收到的 MAC列表添加在自身的 MAC地址转发表中,然后将其端口设置为 PW2。 另一种选择方式是由 PE3直接将端口表项中的 PW1替换为 PW2,然后将替换后的整个 MAC 地址转发表发送给 PE4, 相应地, 步骤 204中的 PE4收到 PE1发送的表项后, 直接保存该 MAC地址转发表;其它步骤不变。或者,由 PE3直接将自身的 MAC地址转发表发送给 PE4, 并指明主 PW和备用 PW的标识信息, 即 PW1和 PW2, PE4收到 MAC地址转发表后, 将 端口为 PW1的表项替换为端口为 PW2。
本实施例在从 PW1切换到 PW2后, PE设备通过 MAC地址转发表的迁移或切换操作 更新了自身的 MAC地址转发表, 进而 PE设备通过 PW2转发报文, 避免了 MAC地址的回 收和重学习过程及其引起的分组洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表 的更新效率, 节省了核心网络的带宽。 实施例 3
本实施例提供了一种以太网业务传送中转发表的更新方法, 参见图 8 提供的组网结构 示意图, 其中, PE1、 PE2、 PW1、 PW2中的功能同图 7中相同, 本实施例以图 8中的 PE1 发生故障, 由 PW1切换到 PW2为例进行说明, MAC地址转发表的更新方法包括:
PE2禾口 PE3检测至 lj PE1发生故障后, 将 PW1切换至 lj PW2, S卩 PE2激活 PW2, PE3进 行转发表的迁移操作;
其中, PE3将会按照实施例 2提供的 MAC地址转发表的迁移操作,完成 PE4上的 MAC 地址转发表的更新, 这里不再详述;
PE2激活 PW2后, 直接可以应用自身预先保存的 MAC地址转发表进行报文转发, 该 MAC地址转发表为 PE1正常工作时,定期将自身的对应于 PW1的 MAC地址转发表向 PE2 按上述过程迁移而得到。
如果 PE2检测到 PE1发生故障时,本身还没有从 PE1处得到 MAC地址转发表,即 PE2 自身的 MAC地址转发表为空, 则需要通过自学习的方式获取 MAC地址转发表。
本实施例在 PE1故障后, 从 PW1切换到 PW2后, PE2直接启用自身的 MAC地址转发 表, PE3通过 MAC地址转发表迁移操作使 PE4更新其 MAC地址转发表, 然后 PE2和 PE4 通过 PW2转发报文, 避免了 MAC地址的回收和重学习过程及其引起的分组洪泛, 减少了 对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的带宽。 实施例 4
本实施例提供了一种以太网业务传送中转发表的更新方法, 本实施例以网络中的一个 指定设备 (例如, 管理平台) 进行 PW切换的判断和决策为例进行说明, 例如, 管理平台 可以进行无故障下的人工 PW切换操作, 也可以在收集到故障消息 (例如 AC PW或 PE 设备失效时) 后再进行 PW切换的响应。 该指定设备负责通知对应的 PE设备进行 MAC地 址转发表迁移或进行 MAC地址转发表切换操作, 对应的 PE设备将会根据指定设备的通知 执行 MAC地址转发表迁移或进行 MAC地址转发表切换操作, 具体的 MAC地址转发表迁 移或 MAC地址转发表切换操作同上述实施例 1或 2类似。
下面以图 9提供的组网结构示意图为例,简单描述一下 MAC地址转发表迁移和切换过 程, 其中, PE1 PE2分别是主 PE设备和备用 PE设备, PW1为 PW3的主 PW PW2为 PW4的主 PW, 进行 PW切换前, PE1上的 MAC地址转发表如表 1所示, PE3上的 MAC 地址转发表如表 2所示, PE4上的 MAC地址转发表如表 3所示 (为便于理解, 增加了源 MAC地址列):
表 1
Figure imgf000012_0001
表 3
Figure imgf000013_0001
当检测到 AC1发生故障后, 通过特定设备 (例如网络管理设备) 判断 PW的切换是否 引起本地 PE设备发生改变, 判断出 PW切换后, 对于 PE1来说, 导致本地 PE设备发生改 变, 即由原来的 PE1变为 PE2, 对于 PE3和 PE4来说, 本地 PE设备没有发生改变, 所以, 向 PE1发送 MAC地址转发表迁移通知, 向 PE3和 PE4发送 MAC地址转发表切换通知; 按照上述实施例 2提供的 MAC地址转发表的迁移方法, PE1收到指定设备发送的 MAC 地址转发表迁移通知后, PE1向 PE2发送 MAC地址转发表迁移消息 Mac_Transfer(PWl, PW3: MAC Listl), Mac_Transfer(PW2, PW4, MAC_List2), 其中 PW1 PW2禾 P PW3 PW4分别 为主 PW和备用 PW的标识, MAC_Listl为 PE1的 MAC地址转发表中端口为 PW1对应的 部分表项或全部表项, MAC_List2为 PE1的 MAC地址转发表中端口为 PW2对应的部分表 项或全部表项, 此外, 这两条 MAC迁移消息也可以封装在同一个 LDP消息中进行传送;
PE2收到 PE1发送的 MAC地址转发表迁移消息后, 将 MAC地址列表添加到自身的 MAC地址转发表中, 并将这些 MAC地址对应的端口分别由 PW1替换为 PW3 PW2替换 为 PW4;
按照上述实施例 1提供的 MAC地址转发表的切换方法, PE3和 PE4主动进行 PW切换 或者收到并执行指定设备发送的 MAC地址转发表切换通知后, PE3将自身的 MAC地址转 发表中的 PW1替换为 PW3 PE4将自身的 MAC地址转发表中的 PW2替换为 PW4
参见图 10, 完成上述操作后, PE2上的 MAC地址转发表如表 4所示, PE3上的 MAC 地址转发表如表 5所示, PE4上的 MAC地址转发表如表 6所示(为便于理解, 增加了源地 址列):
表 4
Figure imgf000013_0002
表 5
Figure imgf000014_0001
上述 PE1到 PE2中的 MAC地址转发表的更新过程为 MAC地址转发表的迁移过程, PE3 和 PE4的 MAC地址转发表的更新过程为 MAC地址转发表的切换过程, MAC地址转发表 迁移和切换过程的具体实现与实施例 1和实施例 2类似, 这里不再详细描述。
本实施例通过指定设备通知对应的 PE设备进行 MAC地址转发表迁移或切换过程, 进 而实现了 MAC地址转发表的更新, 避免了 MAC地址的回收和重学习过程及其引起的分组 洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的 带宽。
上述实施例 1-4中的 VPLS实例中可能有多个 AC接入同一个 PE设备, 并通过同一个 PW进行数据报文传送; 另外在负载分担的时候, 来自同一个 AC的以太网业务可能通过多 个 PW或分组交换路径进行数据报文传送。 此时可能只需要将对应于特定 AC的部分 MAC 地址表进行切换或迁移, 这可通过发送该部分 MAC地址列表来实现, 该 MAC地址列表类 似于现有技术中 RFC 4762中定义的 MAC回收消息中的 MAC地址列表。若需要将对应 PW 的 MAC 转发表全部切换则发送一个空的 MAC 列表。 此时转发表更新通知可以采用 MAC_Switch (PWl, PW2, Mac List) 消息。 其中 PE1 PE2分别是主 PE设备和备用 PE设 备, PW1 PW2分别是主 PW和备用 PW Mac_List为 MAC列表。 以图 7中的 PE1为例, 如果 PE1连接多个 CE设备, 有的 CE设备需要保护, 例如图 7中的 CE1 ; 有的 CE设备不 需要保护, 此时, PE1从自身的 MAC地址转发表中找出端口为 PW1对应的表项后, 只需 将那些在 PE2上进行保护的 CE设备的 MAC地址对应的 MAC地址表项发送给 PE2 上述 MAC地址转发表迁移和切换都可以分为主动模式和被动模式。在主动模式下, PE 设备主动进行 MAC地址转发表迁移和切换,这种设备通常为 PW冗余切换独立模式下的发 起方或主从模式下的主 PE设备; 在被动模式下, PE设备在接收到转发表迁移或切换通知 消息之后才进行转发表迁移或切换, 在接收到该通知后, PE设备将转发表中的主 PW替换 为备用 PW。 实施例 5
参见图 11, 本实施例提供了一种运营商边缘设备, 包括:
故障处理模块 301, 用于检测链路或 PE设备出现故障时, 将主链路切换到备用链路; 转发表更新模块 302, 用于故障处理模块 301 从主链路切换到备用链路时, 将自身的
MAC地址转发表中的与主链路对应的端口更新为与备用链路对应的端口。
本实施例中的端口既可以是物理端口, 也可以是逻辑端口。
本发明实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包括 PW等。 本实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪线本身既 可以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理链路、 端 口为物理端口的情况, 本实施例提供的技术方案仍然适用。 本实施例中的运营商边缘设备 具体进行 MAC地址转发表更新时, 首先检查自身的 MAC地址转发表, 从 MAC地址转发 表中找出端口为主伪线的每个对应表项, 然后将各表项中的主伪线替换为备用伪线。 这里 不再详述。
本实施例中的运营商边缘设备进行主链路切换到备用链路时,将自身的 MAC地址转发 表中的与主链路对应的端口更新为备用链路对应的端口; 并通过备用链路转发报文, 避免 了 MAC地址的回收和重学习过程及其引起的分组洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的带宽。 实施例 6
参见图 12, 本实施例提供了一种运营商边缘设备, 包括:
判断模块 401, 用于主链路切换到备用链路时, 判断备用链路上对应的 PE设备是否为 自身;
处理模块 402,用于当判断模块 401的判断结果是备用链路上对应的 PE设备为自身时, 将自身的 MAC地址转发表中的与主链路对应的端口替换为与备用链路对应的端口; 否则, 向备用链路上对应的 PE设备发送 MAC地址转发表迁移消息; 该消息可以包含 MAC地址 列表、 主链路和备用链路的标识。
本实施例中的端口既可以是物理端口, 也可以是逻辑端口。
本实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包 括 PW等。 本实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪线本身既可 以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理链路、 端口 为物理端口的情况, 本实施例提供的技术方案仍然适用。
本实施例中的运营商边缘设备具体进行 MAC地址转发表更新时,可以采用类似实施例 1、 2、 3或 4中的 PE3使用的更新方法, 即备用伪线上的本地 PE设备为自身时, 检查自身 的 MAC地址转发表, 从 MAC地址转发表中找出端口为主伪线的每个对应表项, 然后将各 表项中的主伪线替换为备用伪线; 备用伪线上的本地 PE设备不是自身时, 发起 MAC地址 转发表的迁移操作, 这里不再详述。
本实施例中的运营商边缘设备根据判断模块 401的判断结果决定执行 MAC地址转发表 的切换或迁移操作, 避免了 MAC地址的回收和重学习过程及其引起的分组洪泛, 减少了对 网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的带宽。 实施例 7
参见图 13, 本实施例提供了一种运营商边缘设备, 包括:
接收模块 501, 用于接收主链路上对应的 PE设备发送的 MAC地址转发表迁移消息; 处理模块 502, 用于接收模块 501收到 MAC地址转发表迁移消息后, 将 MAC地址转 发表迁移消息中包含的 MAC地址加入自身的 MAC地址转发表, 并将这些 MAC地址的端 口设置为与备用链路对应的端口。
本实施例中的端口既可以是物理端口, 也可以是逻辑端口。
本实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包 括 PW等。 本实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪线本身既可 以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理链路、 端口 为物理端口的情况, 本实施例提供的技术方案仍然适用。
本实施例中的运营商边缘设备为备用运营商边缘设备, 进行 MAC地址转发表更新时, 根据 MAC地址转发表迁移消息更新自身的 MAC地址转发表, 避免了 MAC地址学习过程 及其引起的分组洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节 省了核心网络的带宽。 实施例 8
参见图 14, 本实施例提供了一种以太网业务传送中转发表的更新系统, 包括: 主运营商边缘设备 601, 用于主链路切换到备用链路时, 判断备用链路上的备用运营商 边缘设备 602是否为自身; 如果是, 将自身的 MAC地址转发表中的与主链路对应的端口替 换为与备用链路对应的端口; 否则, 向备用运营商边缘设备 602发送 MAC地址转发表迁移 消息;
备用运营商边缘设备 602, 用于收到主运营商边缘设备 601的 MAC地址转发表迁移消 息后, 将 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC地址转发表, 并 将这些 MAC地址的端口设置为与备用链路对应的端口。
本实施例中的端口既可以是物理端口, 也可以是逻辑端口。
本实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包 括 PW等。 本实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪线本身既可 以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理链路、 端口 为物理端口的情况, 本实施例提供的技术方案仍然适用。
主运营商边缘设备 601和备用运营商边缘设备 602进行 MAC地址转发表更新的具体操 作类似于实施例 2中的 PE1和 PE2进行更新的操作, 这里不再详述。
本实施例中的系统通过主运营商边缘设备 601和备用运营商边缘设备 602进行 MAC地 址转发表的更新, 将其中的主 PW更新为备用 PW, 避免了 MAC地址的回收和重学习过程 及其引起的分组洪泛, 减少了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节 省了核心网络的带宽。 实施例 9
参见图 15, 本实施例提供了一种以太网业务传送中转发表的更新系统, 包括: 判断装置 701, 用于主链路切换到备用链路时, 判断主链路切换到备用链路是否导致对 应的 PE设备发生变化, 如果是, 向主链路上的主运营商边缘设备 702发送 MAC地址转发 表迁移通知; 否则, 向主链路上的主运营商边缘设备 702发送 MAC地址转发表切换通知; 主运营商边缘设备 702, 用于收到判断装置 701发送的 MAC地址转发表迁移通知后, 向备用链路上的备用运营商边缘设备 703发送 MAC地址转发表迁移消息; 以及收到判断装 置 701发送的 MAC地址转发表切换通知后, 将自身的 MAC地址转发表中的与主链路对应 的端口替换为与备用链路对应的端口;
备用运营商边缘设备 703, 用于收到主运营商边缘设备 702的 MAC地址转发表迁移消 息后, 将 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC地址转发表, 并 将这些 MAC地址的端口设置为与备用链路对应的端口。
本实施例中的端口既可以是物理端口, 也可以是逻辑端口。
本实施例中的链路可以是物理链路, 也可以是逻辑链路, 其中, 逻辑链路有多种, 包 括 PW等。 本实施例以主、 备链路为伪线, 端口为逻辑端口 (即伪线, 因为伪线本身既可 以是一种逻辑链路, 又可以是一种逻辑端口) 为例进行说明, 对于链路为物理链路、 端口 为物理端口的情况, 本实施例提供的技术方案仍然适用。
上述系统进行 MAC地址转发表更新的具体操作类似于实施例 4的图 9中的设备进行的 更新操作, 其中, 判断装置具有实施例 4中的指定设备的功能, 这里不再详述。
本实施例中的系统通过判断装置 701触发主运营商边缘设备 702进行 MAC地址转发表 的迁移操作或切换操作, 避免了 MAC地址的回收和重学习过程及其引起的分组洪泛, 减少 了对网络业务的影响, 提高了 MAC地址转发表的更新效率, 节省了核心网络的带宽。 在通常部署的 VPLS网络下, 各个 PE之间一般采用全栅格的连接, 也就是说在每一对 PE之间, 只存在一条伪线。 因此, 以上实施例中的所有消息通告, 也可以携带主节点和备 用节点的标识信息以及 MAC地址列表,接收的运营商边缘设备则根据主节点和备用节点的 标识信息, 即可取得它们所对应的主链路和备用链路, 并按照以上实施例方法进行 MAC地 址更新。 另外, 由于根据消息所在的信道或者发送消息的节点本身也足以确定链路或者节 点的信息, 因此所述消息在优化情况下只需要携带一个节点或一条链路的信息即可。 以上实施例提供的技术方案可以应用于具有冗余架构的组网情况, 例如主备保护或负 载分担情况下的组网结构, 利用 PE设备和 PW群组的冗余保护关系, 可避免以太网多点业 务传送下的 MAC地址回收和自学习过程, 以便减少通信延迟, 避免 MAC地址自学习过程 时带来的过多带宽消耗, 增加以太网多点业务的可靠性和可扩展性。
上述实施例以 VPLS作为具体实例进行描述的,但本发明实施例提供的技术方案可应用 于其它分组传送网络技术, 例如 PBBN 禾 P PBB-TE ( Provider Backbone Bridge-Traffic Engineering, 运营商骨干网桥一流量工程) 等网络中。
以上实施例提供的技术方案中的全部或部分内容可以通过软件编程实现, 其软件程序 存储在可读取的存储介质中, 存储介质例如: 计算机中的硬盘、 光盘或软盘。
以上仅为本发明的较佳实施例, 并不用以限制本发明, 凡在本发明的精神和原则之内, 所 作的任何修改、 等同替换、 改进等, 均应包含在本发明的保护范围之内。

Claims

权 利 要 求 书
1. 一种以太网业务传送中转发表的更新方法, 其特征在于, 所述方法包括:
当主链路切换到备用链路时, 运营商边缘 (PE) 设备根据所述主链路、 所述备用链路以 及媒体访问控制 (MAC ) 地址列表更新自身的 MAC地址转发表。
2. 如权利要求 1所述的以太网业务传送中转发表的更新方法, 其特征在于, 所述运营商 边缘(PE)设备根据所述主链路、 所述备用链路以及媒体访问控制 (MAC )地址列表更新自 身的 MAC地址转发表包括:
所述 PE设备接收包含 MAC地址列表的消息;
所述 PE设备收到所述消息后, 根据所述 MAC地址列表更新自身的 MAC地址转发表。
3. 如权利要求 2所述的以太网业务传送中转发表的更新方法, 其特征在于, 所述 PE设 备收到所述 MAC地址列表后,根据所述 MAC地址列表更新自身的 MAC地址转发表,包括: 若所述 MAC地址列表非空, 则所述 PE设备将自身的 MAC地址转发表中对应于所述 MAC地址列表中的 MAC地址表项的端口设置为与所述备用链路对应的端口;
若所述 MAC地址列表为空, 则所述 PE设备将自身的 MAC地址转发表中与所述主链路 对应的所有表项的端口更新为与所述备用链路对应的端口。
4. 如权利要求 2所述的以太网业务传送中转发表的更新方法, 其特征在于, 所述 PE设 备收到所述 MAC地址列表后,根据所述 MAC地址列表更新自身的 MAC地址转发表,包括: 所述 PE设备将所述 MAC地址列表中的 MAC地址加入自身的 MAC地址转发表中, 并 将所述 MAC地址对应表项的端口设置为与所述备用链路对应的端口。
5. 如权利要求 1所述的以太网业务传送中转发表的更新方法, 其特征在于, 运营商边缘 (PE)设备根据所述主链路、 所述备用链路以及媒体访问控制(MAC )地址列表更新自身的
MAC地址转发表包括:
所述主链路上的 PE设备判断自身是否为所述备用链路上的 PE设备;
如果是, 则将自身的 MAC地址转发表中的与所述主链路对应的所有表项的端口更新为 与所述备用链路对应的端口;
否则, 向所述备用链路上本地对应的 PE设备发送 MAC地址转发表迁移消息, 所述消息 包含所述主链路上的 PE设备的 MAC地址转发表中与所述主链路对应的 MAC地址列表; 所述备用链路上本地对应的 PE设备收到所述 MAC地址转发表迁移消息后,将所述 MAC 地址列表中的 MAC地址加入自身的 MAC地址转发表中, 并将所述 MAC地址对应表项的端 口设置为与所述备用链路对应的端口。
6. 如权利要求 1所述的以太网业务传送中转发表的更新方法, 其特征在于, 运营商边缘 (PE)设备根据所述主链路、 所述备用链路以及媒体访问控制(MAC)地址列表更新自身的
MAC地址转发表包括:
指定设备判断所述主链路切换到所述备用链路是否导致对应的 PE 设备发生变化, 如果 是, 向所述主链路上的 PE设备发送 MAC地址转发表迁移通知; 否则, 向所述主链路上的 PE设备发送 MAC地址转发表切换通知;
所述主链路上的 PE设备收到所述 MAC地址转发表迁移通知后, 向所述备用链路上对应 的 PE设备发送 MAC地址转发表迁移消息,所述消息包含所述主链路对应的 MAC地址列表; 所述备用链路上本地对应的 PE设备将所述 MAC地址列表中的 MAC地址加入自身的 MAC 地址转发表中, 将所述 MAC地址的端口设置为与所述备用链路对应的端口;
所述主链路上的 PE设备收到所述 MAC地址转发表切换通知后, 将自身的 MAC地址转 发表中的与所述主链路对应表项的端口替换为与所述备用链路对应的端口。
7. 如权利要求 1所述的以太网业务传送中转发表的更新方法, 其特征在于, 运营商边缘 (PE)设备根据所述主链路、 所述备用链路以及媒体访问控制(MAC)地址列表更新自身的
MAC地址转发表包括:
所述备用链路上的 PE设备将自身的 MAC地址转发表中的所述主链路对应表项的端口更 新为所述备用链路对应的端口后, 向所述备用链路上对端的 PE 设备发送转发表更新通知消 息;
所述备用链路上对端的 PE设备收到所述转发表更新通知消息后,将自身的 MAC地址转 发表中的与所述主链路对应表项的端口更新为与所述备用链路对应的端口。
8. 如权利要求 2至 7中任一权利要求所述的以太网业务传送中转发表的更新方法, 其特 征在于, 所述消息进一步包含下列信息之一: 主链路标识、 备用链路标识、 主节点标识、 从 节点标识。
9. 如权利要求 1至 8中任一权利要求所述的以太网业务传送中转发表的更新方法, 其特 征在于, 所述方法还包括:
所述 PE设备根据更新后的 MAC地址转发表进行报文转发。
10. 如权利要求 1至 8中任一权利要求所述的以太网业务传送中转发表的更新方法, 其 特征在于, 所述链路为逻辑链路或伪线, 相应地, 所述端口为逻辑端口。
11.一种运营商边缘设备, 其特征在于, 所述设备包括:
故障处理模块 (301 ), 用于链路或运营商边缘 (PE) 设备出现故障时, 将主链路切换到 备用链路;
转发表更新模块(302), 用于所述故障处理模块(301 )从主链路切换到备用链路时, 根 据所述主链路、 所述备用链路以及媒体访问控制 (MAC) 地址列表更新自身的 MAC地址转 发表。
12.—种运营商边缘设备, 其特征在于, 所述设备包括:
判断模块 (401 ), 用于主链路切换到备用链路时, 判断所述备用链路是否有一端运营商 边缘 (PE) 设备为自身;
处理模块 (402), 用于当所述判断模块 (401 ) 的判断结果是所述备用链路有一端 PE设 备为自身时, 将自身的媒体访问控制 (MAC) 地址转发表中的与所述主链路对应的表项的端 口替换为与所述备用链路对应的端口;否则,向所述备用链路上本地对应的 PE设备发送 MAC 地址转发表迁移消息。
13.—种运营商边缘设备, 其特征在于, 所述设备包括:
接收模块 (501 ), 用于接收包含媒体访问控制 (MAC) 地址列表的消息;
处理模块 (502), 用于所述接收模块 (501 ) 收到所述消息后, 根据所述 MAC地址列表 中包含的 MAC地址更新自身的 MAC地址转发表, 并将所述 MAC地址的端口设置为与所述 备用链路对应的端口。
14. 一种以太网业务传送中转发表的更新系统, 其特征在于, 所述系统包括: 主运营商边缘设备 (601 ), 用于主链路切换到备用链路时, 判断所述备用链路是否有一 端运营商边缘设备 (PE) 为自身; 如果是, 将自身的媒体访问控制 (MAC)地址转发表中的 与所述主链路对应表项的端口替换为与所述备用链路对应的端口; 否则, 向备用运营商边缘 设备发送 MAC地址转发表迁移消息;
备用运营商边缘设备 (602), 用于收到所述主运营商边缘设备 (601 ) 的 MAC地址转发 表迁移消息后, 将所述 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC地址 转发表中, 并将所述 MAC地址表项的端口设置为与所述备用链路对应的端口。
15. 一种以太网业务传送中转发表的更新系统, 其特征在于, 所述系统包括:
判断装置 (701 ), 用于主链路切换到备用链路时, 判断所述主链路切换到所述备用链路 是否导致运营商边缘 (PE)设备发生变化,如果是,向所述主链路上的主运营商边缘设备(702) 发送媒体访问控制 (MAC)地址转发表迁移通知; 否则, 向所述主链路上的主运营商边缘设 备 (702) 发送 MAC地址转发表切换通知;
主运营商边缘设备 (702), 用于收到所述判断装置发送的 MAC地址转发表迁移通知后, 向所述备用链路上的对应备用运营商边缘设备 (703 ) 发送 MAC地址转发表迁移消息; 以及 收到所述判断装置 (701 ) 发送的 MAC地址转发表切换通知后, 将自身的 MAC地址转发表 中的与所述主链路对应表项的端口替换为与所述备用链路对应的端口;
备用运营商边缘设备 (703 ), 用于收到所述主运营商边缘设备 (702) 的 MAC地址转发 表迁移消息后, 将所述 MAC地址转发表迁移消息中包含的 MAC地址加入自身的 MAC地址 转发表中, 并将所述 MAC地址的表项端口设置为与所述备用链路对应的端口。
PCT/CN2009/071782 2008-05-16 2009-05-13 以太网业务传送中转发表的更新方法、设备和系统 WO2009138036A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810081885.7 2008-05-16
CN200810081885.7A CN101582834B (zh) 2008-05-16 2008-05-16 以太网业务传送中转发表的更新方法和系统

Publications (1)

Publication Number Publication Date
WO2009138036A1 true WO2009138036A1 (zh) 2009-11-19

Family

ID=41318371

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/071782 WO2009138036A1 (zh) 2008-05-16 2009-05-13 以太网业务传送中转发表的更新方法、设备和系统

Country Status (2)

Country Link
CN (1) CN101582834B (zh)
WO (1) WO2009138036A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013174327A1 (zh) * 2012-08-31 2013-11-28 中兴通讯股份有限公司 一种媒体流传输方法及装置
CN104702435A (zh) * 2015-02-05 2015-06-10 盛科网络(苏州)有限公司 网络处理器实现aps切换的方法及网络处理器
CN105141521A (zh) * 2015-09-21 2015-12-09 盛科网络(苏州)有限公司 使用保护切换机制实现多系统链路聚合的装置及方法
CN110457328A (zh) * 2019-08-15 2019-11-15 中国银行股份有限公司 参数生效方法及装置

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102148677B (zh) * 2010-02-10 2013-06-26 杭州华三通信技术有限公司 一种更新地址解析协议表项的方法及核心交换机
CN101873246B (zh) * 2010-06-11 2012-09-12 福建星网锐捷网络有限公司 一种控制路由设备接口状态的方法及装置
CN101877677B (zh) * 2010-06-25 2014-08-13 中兴通讯股份有限公司 一种多协议标签交换业务隧道切换的方法及系统
CN101977148B (zh) * 2010-10-26 2015-01-28 中兴通讯股份有限公司 内容分发网络节点媒体服务器之间数据交互的方法及系统
CN102025541B (zh) * 2010-12-08 2014-12-10 中兴通讯股份有限公司 一种实现组播保护的方法及系统
CN102546357B (zh) * 2010-12-09 2014-12-17 华为技术有限公司 报文转发方法及网络设备
CN102123085B (zh) * 2011-02-16 2014-03-19 杭州华三通信技术有限公司 上行链路的切换控制方法以及切换控制装置
CN102694718B (zh) * 2011-03-25 2016-03-30 华为技术有限公司 一种vpls快速重路由方法和设备
CN102546855B (zh) * 2011-11-28 2017-10-17 中兴通讯股份有限公司 Vpls网络中mac地址回收方法及pe设备
CN102571426B (zh) 2011-12-29 2015-05-20 杭州华三通信技术有限公司 一种双归保护方法和装置
CN103795634B (zh) * 2012-11-01 2019-02-05 中兴通讯股份有限公司 一种转发表的生成方法及装置
CN103825815B (zh) * 2012-11-16 2018-07-27 中兴通讯股份有限公司 网络虚拟边界设备间进行冗余备份的方法、设备及系统
CN103067243B (zh) * 2012-12-21 2016-03-09 华为技术有限公司 通信方法及相关设备
CN104283778B (zh) * 2013-07-10 2017-11-14 新华三技术有限公司 最短路径桥网络内的流量保护方法及设备
CN103414642B (zh) * 2013-08-19 2016-11-02 杭州华三通信技术有限公司 一种转发信息维护方法及设备
CN105900406B (zh) * 2013-12-23 2019-07-09 瑞典爱立信有限公司 针对网络服务可用性的技术
CN105577417B (zh) * 2014-11-06 2019-02-22 新华三技术有限公司 基于vxlan网络的报文转发方法及装置
CN105656775A (zh) * 2014-11-14 2016-06-08 中兴通讯股份有限公司 一种介质访问控制转发链路切换方法及装置
CN105991783A (zh) * 2015-02-06 2016-10-05 中兴通讯股份有限公司 一种媒体接入控制地址学习方法、设备和系统
CN105450518B (zh) * 2015-12-04 2018-08-03 福建星网锐捷网络有限公司 一种mpls-tp环网故障排除方法及装置
CN108075968A (zh) * 2016-11-10 2018-05-25 中国移动通信集团广东有限公司 一种网络系统及伪线业务处理方法
CN107872391B (zh) * 2017-11-23 2021-06-18 锐捷网络股份有限公司 一种表项更新方法及装置
CN107733724B (zh) * 2017-11-24 2020-12-22 瑞斯康达科技发展股份有限公司 一种环形网络的配置方法、装置及环形网络
CN108199962B (zh) * 2017-12-22 2021-09-07 新华三技术有限公司 地址迁移方法、装置、网络设备及可读存储介质
CN110740094B (zh) * 2019-10-28 2021-09-21 新华三信息安全技术有限公司 一种网络设备、bfd报文的传输方法及装置
CN111092822A (zh) * 2019-12-25 2020-05-01 安徽皖通邮电股份有限公司 一种vpls的pw线性保护转发负载均衡的方法
CN111478851B (zh) * 2020-02-29 2022-05-27 新华三信息安全技术有限公司 一种报文处理方法和装置
CN113726658B (zh) * 2021-08-09 2022-10-18 中国联合网络通信集团有限公司 一种路由转发方法及装置
CN113794657B (zh) * 2021-09-14 2023-10-31 迈普通信技术股份有限公司 Mac地址迁移处理方法、装置及交换设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1909501A (zh) * 2005-08-05 2007-02-07 华为技术有限公司 一种端到端业务快速收敛的方法和路由设备
WO2007066910A1 (en) * 2005-12-09 2007-06-14 Electronics And Telecommunications Research Institute Apparatus and method for multi-protocol label switching label-switched path protection switching
CN101035047A (zh) * 2007-04-03 2007-09-12 中兴通讯股份有限公司 环网保护的处理方法
CN101123531A (zh) * 2006-12-30 2008-02-13 华为技术有限公司 基于vpls双归属的全连接网络的收敛方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1909501A (zh) * 2005-08-05 2007-02-07 华为技术有限公司 一种端到端业务快速收敛的方法和路由设备
WO2007066910A1 (en) * 2005-12-09 2007-06-14 Electronics And Telecommunications Research Institute Apparatus and method for multi-protocol label switching label-switched path protection switching
CN101123531A (zh) * 2006-12-30 2008-02-13 华为技术有限公司 基于vpls双归属的全连接网络的收敛方法和装置
CN101035047A (zh) * 2007-04-03 2007-09-12 中兴通讯股份有限公司 环网保护的处理方法

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013174327A1 (zh) * 2012-08-31 2013-11-28 中兴通讯股份有限公司 一种媒体流传输方法及装置
CN104702435A (zh) * 2015-02-05 2015-06-10 盛科网络(苏州)有限公司 网络处理器实现aps切换的方法及网络处理器
CN104702435B (zh) * 2015-02-05 2018-08-14 盛科网络(苏州)有限公司 网络处理器实现aps切换的方法及网络处理器
CN105141521A (zh) * 2015-09-21 2015-12-09 盛科网络(苏州)有限公司 使用保护切换机制实现多系统链路聚合的装置及方法
CN110457328A (zh) * 2019-08-15 2019-11-15 中国银行股份有限公司 参数生效方法及装置
CN110457328B (zh) * 2019-08-15 2022-02-25 中国银行股份有限公司 参数生效方法及装置

Also Published As

Publication number Publication date
CN101582834B (zh) 2013-10-09
CN101582834A (zh) 2009-11-18

Similar Documents

Publication Publication Date Title
WO2009138036A1 (zh) 以太网业务传送中转发表的更新方法、设备和系统
CN105743689B (zh) 对多宿主以太网虚拟专网中的链路故障的快速收敛
US9781032B1 (en) MPLS label usage in ethernet virtual private networks
US9019814B1 (en) Fast failover in multi-homed ethernet virtual private networks
WO2018054156A1 (zh) 一种vxlan报文的转发方法、设备及系统
US8331220B2 (en) Edge node redundant system
US9509591B2 (en) Technique for dual homing interconnection between communication networks
WO2012028029A1 (zh) 一种切换方法及系统
WO2007062559A1 (fr) Procede et passerelle de restitution de service au moment de la permutation entre passerelles pilote et asservie
US20130272114A1 (en) Pseudo wire switching method and device
WO2008083590A1 (fr) Procédé et appareil de convergence rapide d'un service point à point
WO2006017982A1 (fr) Procede de reacheminement dans le reseau multiprotocole a commutateur d'etiquettes
WO2007016834A1 (fr) Procede rapide de convergence de services de point a point et dispositif associe cote fournisseur de services
JP5168230B2 (ja) 通信システム、エッジルータ及び信号転送方法
CN111064596B (zh) 对于用于多宿主节点故障的bum流量的节点保护
WO2007012239A1 (fr) Procédé permettant de commuter la prestation de services d'un lan privé virtuel et système y afférant
WO2009056034A1 (fr) Procédé, système et équipement pour établir une détection bfd pour un tunnel lsp
CN112422307B (zh) Evpn和vpls共存双活的方法、设备及系统
WO2012109941A1 (zh) 一种trill网络的冗余备份方法及系统
US11349749B2 (en) Node protection for bum traffic for multi-homed node failure
WO2011060667A1 (zh) 一种虚拟专用局域网络中链路保护的方法及设备
WO2012130034A1 (zh) 一种vpls快速重路由方法和设备
WO2012159489A1 (zh) 一种伪线双归网络的切换方法、系统和双归属运营商设备
WO2008028382A1 (fr) Procédé et appareil pour exécuter une détection de liaison, conversion de stratégie d'acheminement de bout en bout
CN102282805B (zh) 一种业务保护方法及接入设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09745420

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09745420

Country of ref document: EP

Kind code of ref document: A1