WO2016082457A1 - 一种媒体接入控制地址收敛方法和装置、存储介质 - Google Patents

一种媒体接入控制地址收敛方法和装置、存储介质 Download PDF

Info

Publication number
WO2016082457A1
WO2016082457A1 PCT/CN2015/077685 CN2015077685W WO2016082457A1 WO 2016082457 A1 WO2016082457 A1 WO 2016082457A1 CN 2015077685 W CN2015077685 W CN 2015077685W WO 2016082457 A1 WO2016082457 A1 WO 2016082457A1
Authority
WO
WIPO (PCT)
Prior art keywords
macwithdraw
event
npe
vpls
mac address
Prior art date
Application number
PCT/CN2015/077685
Other languages
English (en)
French (fr)
Inventor
朱春
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016082457A1 publication Critical patent/WO2016082457A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks

Definitions

  • the present invention relates to a data communication technology, and in particular, to a media access control (MAC) address convergence method and apparatus, and a storage medium in a virtual private LAN service (VPLS) in a Kompella mode.
  • MAC media access control
  • VPLS virtual private LAN service
  • Virtual Private LAN Service is a point-to-multipoint Layer 2 virtual private network (VPN) service provided in a public network.
  • VPN virtual private network
  • VPLS provides Ethernet services to end customers by simulating the carrier's IP core network into a virtual Layer 2 switch.
  • the VPLS network is planned and deployed by the carrier according to the customer and its own requirements.
  • the deployed VPLS network is transparent to the carrier's customers.
  • a typical VPLS network is shown in Figure 1.
  • the VPLS is deployed between the network provider network (NPE, Network Provider Edge) of the IP core network (MPLS): NPE1, NPE2, NPE3, and NPE4 establish a connection through the Hub PW.
  • NPE Network Provider Edge
  • MPLS IP core network
  • NPE1, NPE2, NPE3, and NPE4 establish a connection through the Hub PW.
  • To form a full-mesh VPLS network to avoid loops by restricting the forwarding of packets between Hub PWs.
  • a hierarchical VPLS network model, H-VPLS is proposed.
  • the model is mainly implemented on the access side, and the user-facing provider device (UPE) is used to aggregate the packets of the user-side user edge network device (CE, Custom Edge), and then access the packets through the Spoke PW. Go to the NPE.
  • the active and standby Spoke PWs are dual-homed to different NPEs in the Full-Mesh domain. In this way, when the Lord When the PW (Active PW) fails, the UPE can sense and switch the traffic to the standby PW (Standby PW).
  • the functions of the Layer 2 MAC address learning, flooding, and forwarding on the NPE nodes enable the users in different areas to connect and forward data through the PSN network.
  • Each NPE device is interconnected by a pseudowire (PW, Pseduwire). It is a VPN technology and is the core technology for VPLS service deployment. Through the PW, it is ensured that user traffic accessed by different provider edge devices (PEs) can be directly transmitted transparently on a packet switched network (PSN).
  • PW packet switched network
  • One is the Extended Label Distribution Protocol (LDP) signaling mechanism (also known as Martini mode, RFC4762), and the other is the extended Border Gateway Protocol (BGP). , Border Gateway Protocol) signaling mechanism (also known as Kompella mode, RFC761).
  • LDP Extended Label Distribution Protocol
  • BGP Border Gateway Protocol
  • Border Gateway Protocol also known as Kompella mode, RFC761
  • the advantage of the PW established by using LDP signaling is that the signaling extension is relatively simple and convenient to deploy.
  • a PW that uses BGP signaling can support the dynamic discovery of PEs and the automatic creation of PWs. It is applicable to VPLS services with topology-constrained and numerous nodes, such as the Full-Mesh topology and the H-VPLS core area, which can greatly reduce management costs.
  • the MAC address undo (MacWithdraw) function is to accelerate the convergence of the MAC address by sending a MacWithdraw message when the VPLS forwarding link changes, and to avoid traffic loss in the reverse direction.
  • MacWithdraw MacWithdraw
  • the bidirectional traffic path between CE1 and CE3 is CE1 ⁇ ->UPE ⁇ ->NPE1 ⁇ -> in steady state.
  • the UPE senses and switches the traffic path to NPE2.
  • the traffic path from CE1 to CE3 becomes CE1—>UPE—>NPE2—>NPE3—>CE3.
  • the switch NPE3 cannot be immediately aware, the traffic path from CE3 to CE1 remains the path before the switchover, that is, CE3—>NPE3—>NPE1—>UPE—>CE1. Therefore, when the traffic of CE3 reaches NPE1, The link between NPE1 and UPE fails, causing reverse traffic to be discarded.
  • the industry has proposed the MacWithdraw function. Taking the networking scenario of Figure 1 as an example, when the link between the UPE and the NPE1 fails, the UPE sends a MacWithdraw message to the NPE2. After receiving the MacWithdraw message, the NPE2 continues to forward all the Hub PWs in the Full-Mesh domain.
  • the message at the same time, after receiving the message, clears all previously learned MAC addresses in the VPLS, except for the MAC address learned from the PW that received the MacWithdraw message.
  • NPE3 After receiving the MacWithdraw message, NPE3 retains only the MAC address previously learned from Hub PW23, and deletes all other MAC addresses. In this way, after the traffic of the CE3 reaches the NPE3, the MAC address table of the NPE3 will not go to the path of the NPE1 that has expired before, and then the Hub PW23 is sent to the NPE2 and then sent to the CE1. Therefore, the forwarding link between CE1 and CE3 is consistent, and packet loss is avoided.
  • Section 6.2 of RFC4762 defines the implementation standard for the MacWithdraw function in the VPLS service created by the Martini method. That is, by extending the LDP Address Withdraw message, adding a Mac List TLV to carry the MAC address information that needs to be revoked, when the Mac list in the Mac List TLV is empty, it means deleting all the MAC addresses learned before (except from the MAC address) The MAC address learned on the PW of the MacWithdraw message is received. Otherwise, only the MAC address specified in the Mac List TLV is deleted.
  • RFC4761 does not define the implementation of the MacWithdraw function of the VPLS service created by the Kompella method.
  • the VPLS service created by the Kompella mode is the same as the VPLS created by the Martini mode. Therefore, when the VPLS forwarding link changes, the reverse data traffic will be lost and the traffic will be interrupted. .
  • the present invention mainly provides a MAC address convergence method and apparatus.
  • the present invention provides a media access control MAC address convergence method, and the method includes:
  • the network-oriented provider device NPE establishes the Kompella virtual private LAN service VPLS service.
  • the border gateway protocol BGP message is sent to the other NPE devices in the VPLS domain to cancel the MacWithdraw event.
  • the MAC address deletion operation is performed according to the policy of the MacWithdraw event.
  • the NPE device learns that the VPLS service forwarding link changes, including: the state change of the local VPLS access interface AC, and/or the change of the access pseudowire Spoke PW state, and/or the received The MacWithdraw event sent from the Spoke PW indicating that the Spoke PW status changes from standby to primary Active is known to change the VPLS service forwarding link.
  • the sending the MacWithdraw event to the other NPE devices in the VPLS domain by using the BGP message includes: the NPE device sends the MacWithdraw event to the other NPE devices in the VPLS domain by using the PW by extending the BGP message or newly defining a BGP message.
  • the extended BGP message is an extended BGP update Update message, including: newly defining a BGP extended community attribute for the BGP Update message, which is used to carry the MacWithdraw event;
  • the newly defined BGP message defines a new BGP indication Indicicate message, and the BGP Indicate message carries a MacWithdraw event.
  • the MacWithdraw event carries more than one MAC address, or does not carry any valid MAC address.
  • the other NPE device after receiving the BGP message carrying the MacWithdraw event, performs the MAC address deletion operation according to the policy of the MacWithdraw event, including: after the other NPE device receives the BGP message carrying the MacWithdraw event, it carries a message according to the MacWithdraw event.
  • the above MAC address is deleted directly in the local
  • the strategy of the MAC address carried by the MacWithdraw event; according to the MacWithdraw event does not carry any valid MAC address, the strategy of deleting all the MAC addresses learned from the PW receiving the MacWithdraw event before deleting the local.
  • the present invention provides a MAC address convergence device, which includes: a service establishment module, a learning module, and a sending module;
  • a service establishment module configured to establish a Kompella VPLS service between NPE devices
  • the learning module is configured to learn that the VPLS service forwarding link changes, and notify the sending module;
  • the sending module is configured to send a MacWithdraw event to other NPE devices in the VPLS domain except the NPE device in the VPLS domain, so that other NPE devices receive the BGP message carrying the MacWithdraw event, and then perform the MAC address deletion operation according to the policy of the MacWithdraw event. .
  • the learning module is further configured to change the state of the local VPLS AC interface, and/or receive the Spoke PW state change, and/or receive the sent from the Spoke PW, indicating that the Spoke PW state changes from Standby to Standby.
  • the Active MacWithdraw event learns that the VPLS service forwarding link has changed.
  • the sending module is further configured to use the PW to send a MacWithdraw event to other NPE devices in the VPLS domain by extending the BGP message or newly defining a BGP message.
  • the device further includes a deleting module configured to receive the BGP message carrying the MacWithdraw event sent by the other NPE device, and then carry more than one MAC address according to the MacWithdraw event, and directly delete the MAC address carried by the MacWithdraw event locally.
  • Policy according to the MacWithdraw event does not carry any valid MAC address, the strategy used to delete all the MAC addresses learned from the PW that received the MacWithdraw event before deleting the local.
  • the present invention provides an NPE device including the above-described MAC address convergence device.
  • a storage medium storing a computer program configured to perform the aforementioned medium access control MAC address convergence method.
  • the present invention provides a MAC address convergence method and apparatus.
  • a NPE device learns that a VPLS service forwarding link changes, the NPE device sends a MacWithdraw event to other NPE devices in the VPLS domain through a BGP message.
  • the other NPE devices After the other NPE devices receive the BGP message carrying the MacWithdraw event, perform the MAC address deletion operation according to the policy of the MacWithdraw event.
  • the MAC address on each NPE device is accelerated. Convergence, try to ensure that VPLS reverse traffic does not drop packets.
  • FIG. 1 is a schematic diagram of networking of an existing VPLS network
  • FIG. 2 is a schematic flowchart of a MAC address convergence method according to an embodiment of the present disclosure
  • FIG. 3 is a schematic structural diagram of a MAC address convergence apparatus according to an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram of networking of a VPLS network according to Embodiment 1 of the present invention.
  • FIG. 5 is a schematic diagram of networking of a VPLS network according to Embodiment 2 of the present invention.
  • FIG. 6 is a schematic diagram of networking of a VPLS network according to Embodiment 3 of the present invention.
  • the Kompella VPLS service is established between the NPE devices.
  • the NPE device learns that the VPLS service forwarding link changes, the NPE device sends a MacWithdraw event to other NPE devices in the VPLS domain to enable other NPE devices to receive the packet.
  • the MAC address deletion operation is performed according to the policy of the MacWithdraw event.
  • the embodiment of the invention implements a MAC address convergence method. As shown in FIG. 2, the method includes the following steps:
  • Step 201 Establish a Kompella VPLS service between NPE devices.
  • This step is to establish a Kompella VPLS service between NPE devices based on the signaling procedure of RFC4761.
  • Step 202 The NPE device learns that the VPLS service forwarding link changes.
  • the NPE device may change the state of the local VPLS access interface (AC, Attachment Circuit), and/or receive the Spoke PW state change, and/or receive the Spoke PW status from the Spoke PW. Standby) becomes the active MacWithdraw event to learn that the VPLS service forwarding link has changed.
  • AC Attachment Circuit
  • Step 203 The NPE device sends a MacWithdraw event to other NPE devices in the VPLS domain through the BGP message, so that the other NPE device receives the BGP message carrying the MacWithdraw event, and then performs the MAC address deletion operation according to the policy of the MacWithdraw event.
  • the NPE device may use the PW to send a MacWithdraw event to other NPE devices in the VPLS domain by extending the BGP message or newly defining a BGP message, where the extended BGP message may be an extended BGP update message, for example, to the BGP Update.
  • the message newly defines a BGP Extended Communities Attribute for carrying a MacWithdraw event.
  • the newly defined BGP message may be a newly defined BGP message, such as newly defining a BGP indication message, the BGP.
  • the Indicate message carries the MacWithdraw event;
  • the MacWithdraw event may carry one or more MAC addresses, or may not carry any valid MAC address;
  • the other NPE device After receiving the BGP message carrying the MacWithdraw event, the other NPE device carries one or more MAC addresses according to the MacWithdraw event, and adopts a policy of directly deleting the MAC address carried by the MacWithdraw event locally, according to the MacWithdraw event. Carrying any valid MAC address, adopts the strategy of deleting all the MAC addresses learned from the PW that received the MacWithdraw event before learning locally.
  • the present invention further provides a MAC address convergence device, which is disposed in an NPE device, as shown in FIG. 3, the device includes: a service establishment module 31, a learning module 32, and a sending module 33;
  • the service establishment module 31 can be implemented by an interface of the NPE device, and configured to establish a Kompella VPLS service between the NPE devices.
  • the learning module 32 can be implemented by the detecting chip of the NPE device, configured to learn that the VPLS service forwarding link changes, and notify the sending module 33;
  • the sending module 33 can be implemented by the processor of the NPE device in combination with the interface, and configured to send a MacWithdraw event to other NPE devices in the VPLS domain through BGP messages, so that other NPE devices receive the BGP message carrying the MacWithdraw event, according to the MacWithdraw event.
  • the policy performs a MAC address deletion operation.
  • the learning module 32 may learn the VPLS through a state change of the local VPLS AC interface, and/or receiving a Spoke PW state change, and/or receiving a MacWithdraw event sent from the Spoke PW indicating that the Spoke PW state changes from Standby to Active.
  • the service forwarding link changes.
  • the sending module 33 may use the PW to send a MacWithdraw event to other NPE devices in the VPLS domain by extending the BGP message or newly defining a BGP message, where the extended BGP message may be an extended BGP update message, for example, to the BGP Update.
  • the message newly defines a BGP extended community attribute, which is used to carry a MacWithdraw event.
  • the newly defined BGP message may be a newly defined BGP message, such as newly defining a BGP Indicate message, where the BGP Indicate message carries a MacWithdraw event;
  • the MacWithdraw event may carry one or more MAC addresses, or may not carry any valid MAC address;
  • the device further includes a deleting module 34 configured to receive a BGP message carrying a MacWithdraw event sent by another NPE device, and then carry one or more MAC addresses according to the MacWithdraw event, and adopt a policy of directly deleting the MAC address carried by the MacWithdraw event locally. According to the MacWithdraw event, it does not carry any valid MAC address, and adopts the strategy of deleting all the MAC addresses learned from the PW that received the MacWithdraw event before deleting the local.
  • a deleting module 34 configured to receive a BGP message carrying a MacWithdraw event sent by another NPE device, and then carry one or more MAC addresses according to the MacWithdraw event, and adopt a policy of directly deleting the MAC address carried by the MacWithdraw event locally. According to the MacWithdraw event, it does not carry any valid MAC address, and adopts the strategy of deleting all the MAC addresses learned from the PW that received the MacWithdraw event before deleting the local.
  • the present invention also provides an NPE device, which includes the MAC address convergence device shown in FIG. 3 above.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the scenario of this embodiment is that the dual-homed NPEs in the H-VPLS network are invalid.
  • the UPEs are dual-homed to the NPE1 and NPE2 devices in the Full-Mesh domain through the Spoke PW.
  • the Martini VPLS is deployed.
  • the NPP1, the NPE2, the NPE3, and the NPE4 are deployed with the Kompella VPLS.
  • the forwarding path of the traffic from CE1 to CE3 is CE1->UPE->NPE1->NPE3.
  • the forwarding path of the traffic from CE3 to CE1 is CE3->NPE3->NPE1->UPE->CE1.
  • the UPE1 When the NPE1 fails, the UPE detects that the NPE1 is invalid. The UPE immediately switches the traffic to the Standby Spoke PW. The NPE1 learns the VPLS service forwarding chain based on the traffic switching. The path changes, and an extended BGP Update message is sent to the NPE2 for a MacWithdraw message of an empty media access control list (MacList);
  • MacList media access control list
  • the NPE2 After receiving the MacWithdraw message of the empty MacList sent by the NPE1, the NPE2 deletes all the MAC addresses learned from the Standby Spoke PW before deleting the local VPLS domain, and then passes the extended BGP Update message to the Full-Mesh domain. Other NPE devices, NPE3 and NPE4 (NPE1 has expired, so cannot be sent) continue to send empty MacList MacWithdraw message;
  • the NPE3 After receiving the MacWithdraw message of the empty MacList sent by the NPE2, the NPE3 deletes all the MAC addresses learned from the Hub PW23 before being deleted locally in the VPLS domain;
  • the NPE4 After receiving the MacWithdraw message of the empty MacList sent by the NPE2, the NPE4 deletes all the MAC addresses learned from the Hub PW24 before being deleted locally in the VPLS domain;
  • the original MAC address is cleared.
  • the CE3 sends the traffic to the NPE2 through the broadcast mode.
  • the NPE2 forwards the traffic to the UPE and then forwards it to the CE1.
  • the reverse traffic path from CE3 to CE1 and the traffic path from CE1 to CE3 are the same.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the scenario in this embodiment is that the dual-homed Active Spoke PW in the H-VPLS network is invalid.
  • the UPE is dual-homed to the NPE1 and NPE2 devices in the Full-Mesh domain through the Spoke PW.
  • the Martini VPLS is deployed.
  • the NPP1, the NPE2, the NPE3, and the NPE4 are deployed with the Kompella VPLS.
  • the forwarding path of the traffic from CE1 to CE3 is CE1->UPE->NPE1-> NPE3->CE3;
  • the forwarding path of the traffic from CE3 to CE1 is CE3->NPE3->NPE1->UPE->CE1.
  • both the UPE and the NPE1 learn the change of the VPLS service forwarding link through the PW detection technology or the signaling.
  • the UPE immediately switches the traffic to the Standby Spoke PW, and the NPE1 passes the extended BGP Update message.
  • the MacWithdraw message carrying the MacList is sent directly to all other NPE devices in the Full-Mesh VPLS domain.
  • the MacList contains all the invalid MAC addresses, that is, the MAC addresses learned from the Active Spoke PW.
  • the NPE2 After receiving the MacWithdraw message with the MacList sent by the NPE1, the NPE2 deletes the MAC address carried in the MacList from the local VPLS forwarding table.
  • NPE3 After receiving the MacWithdraw message sent by NPE1 with MacList, NPE3 puts The MAC address carried in the MacList is deleted from the local VPLS forwarding table.
  • the NPE4 After receiving the MacWithdraw message with the MacList sent by the NPE1, the NPE4 deletes the MAC address carried in the MacList from the local VPLS forwarding table.
  • the original MAC address is cleared.
  • the CE3 sends the traffic to the NPE2 through the broadcast mode.
  • the NPE2 forwards the traffic to the UPE and then forwards it to the CE1.
  • the reverse traffic path from CE3 to CE1 and the traffic path from CE1 to CE3 are the same.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • the scenario in this embodiment is that the dual-homed Active AC in the Full-Mesh VPLS domain is invalid.
  • the CE is directly advertised to the NPE1 and NPE2 devices in the Full-Mesh domain.
  • the NPP1, NPE2, NPE3, and NPE4 are deployed with the Kompella VPLS.
  • the forwarding path of the traffic from CE1 to CE2 is CE1->NPE1->NPE3->CE2; the forwarding path of CE3 to CE1 is CE2- >NPE3->NPE1->CE1.
  • both the CE1 and the NPE1 learn the VPLS service forwarding link through the link detection technology.
  • the CE1 immediately switches the traffic to the standby AC.
  • the NPE1 directly forwards the BGP Update message to Full- All other NPE devices in the Mesh VPLS domain send a MacWithdraw message carrying a MacList, which contains all invalid MAC addresses (ie, MAC addresses previously learned from Active AC);
  • the NPE2 After receiving the MacWithdraw message with the MacList sent by the NPE1, the NPE2 deletes the MAC address carried in the MacList from the local VPLS forwarding table.
  • the NPE3 After receiving the MacWithdraw message with the MacList sent by the NPE1, the NPE3 deletes the MAC address carried in the MacList from the local VPLS forwarding table.
  • the NPE4 After receiving the MacWithdraw message with the MacList sent by the NPE1, the NPE4 deletes the MAC address carried in the MacList from the local VPLS forwarding table.
  • CE2 sends the traffic to the NPE2 device through the broadcast mode.
  • the NPE2 device forwards the traffic to CE1.
  • the reverse traffic path from CE2 to CE1 and the traffic path from CE1 to CE2 are the same.
  • the solution provided by the embodiment of the present invention can speed up the convergence of the MAC address on each NPE device when the VPLS forwarding link changes in the Kompella VPLS, and ensure that the VPLS reverse traffic does not lose packets.
  • the embodiment of the invention further describes a storage medium in which a computer program is stored, the computer program being configured to perform the medium access control MAC address convergence method of the foregoing embodiments.
  • the Kompella VPLS service is established between the NPE devices.
  • the NPE device learns that the VPLS service forwarding link changes
  • the BGP message is sent to the other NPE devices in the VPLS domain to send the MacWithdraw event, so that other NPE devices receive the MacWithdraw event.
  • the MAC address deletion operation is performed according to the policy of the MacWithdraw event. In this way, in the Kompella VPLS, when the VPLS forwarding link changes, the MAC address convergence on each NPE device is accelerated, and the VPLS reverse traffic is ensured as much as possible. Do not lose the package.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种媒体接入控制(MAC)地址收敛方法和装置、存储介质，面向网络的提供商设备(NPE)间建立Kompella虚拟专用局域网服务(VPLS)业务，当NPE设备获知VPLS业务转发链路发生变化时，通过边界网关协议(BGP)消息向VPLS域中的其他NPE设备发送MAC地址撤销(MacWithdraw)事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作；本发明同时还公开了一种MAC地址收敛装置。

Description

一种媒体接入控制地址收敛方法和装置、存储介质 技术领域
本发明涉及数据通信技术，尤其涉及一种Kompella方式的虚拟专用局域网服务(VPLS，Virtual Private Lan Service)中媒体接入控制(MAC)地址收敛方法和装置、存储介质。
背景技术
虚拟专用局域网服务(VPLS，Virtual Private Lan Service)是在公用网络中提供的一种点到多点的二层虚拟专用网络(VPN，Virtual Private Network)业务。VPLS通过把运营商的IP核心网模拟成一台虚拟的二层交换机为最终客户提供以太服务。运营商通过部署VPLS保证了分隔在不同地域的相同客户的连通性以及不同用户之间的隔离性。
运营商根据客户以及自身的需求对VPLS网络进行规划和部署，部署的VPLS网络对于运营商的客户是透明的。一个典型的VPLS网络的组网如图1所示。其中，运营商在其IP核心网络(MPLS)的面向网络的提供商设备(NPE，Network Provider Edge)间部署VPLS：NPE1、NPE2、NPE3、NPE4之间通过集线器伪线(Hub PW)建立起连接，构成一个全连接(Full-Mesh)的VPLS网络，通过限制Hub PW之间不能够进行报文转发来避免环路。同时为了分担Full-Mesh网络带来的Hub PW全连接问题，提出了层次化的VPLS网络模型，即H-VPLS。该模型主要体现在接入侧，通过部署面向用户的提供商设备(UPE，User facing-Provider Edge)来汇聚用户侧用户边界网络设备(CE，Custom Edge)的报文，然后通过Spoke PW接入到NPE上。为了保证UPE和NPE之间链路的可靠性，通常在UPE上部署主备Spoke PW双归到Full-Mesh域中不同的NPE设备上。这样，当主 PW(Active PW)发生故障时，UPE可以感知并把流量切换到备PW(Standby PW)上。
当VPLS网络部署完成之后，在转发面，通过各个NPE节点上的二层的MAC地址学习、洪泛、转发等功能使得各个不同区域的用户通过PSN网络进行连接和数据转发。
各个NPE设备之间通过伪线(PW，Pseduwire)进行互连，它是一种VPN技术，是VPLS业务部署的核心技术。通过PW，能够保证不同提供商边缘设备(PE)上接入的用户流量能够在分组交换网(PSN)上直接进行透明传送。PW的建立目前有两种信令机制，一种是扩展的标签分配协议(LDP，Label Distribution Protocol)信令机制(也称为Martini方式，RFC4762)，另外一种是扩展的边界网关协议(BGP，Border Gateway Protocol)信令机制(也称为Kompella方式，RFC761)。使用LDP信令建立的PW的优势是信令扩展比较简单，部署方便。而使用BGP信令建立的PW，能支持PE设备动态发现和PW的自动创建，适用于拓扑规整且节点众多的VPLS服务，比如Full-Mesh拓扑、H-VPLS核心区域，可以大大减少管理成本。
MAC地址撤销(MacWithdraw)功能是在当VPLS转发链路发生变化时，通过发送MacWithdraw消息来加速MAC地址的收敛，避免反方向的流量丢包。举例来说，如图1所示，当UPE和NPE1之间的主PW链路有效的时候，稳定状态下，CE1和CE3之间双向流量路径为CE1<—>UPE<—>NPE1<—>NPE3<—>CE3；当UPE和NPE1之间的主PW链路发生故障时，UPE感知并把流量路径切换到NPE2上。此时，CE1到CE3的流量路径变为了CE1—>UPE—>NPE2—>NPE3—>CE3。但由于此切换NPE3无法立即感知，所以CE3到CE1的流量路径，仍然保持切换前的路径，即CE3—>NPE3—>NPE1—>UPE—>CE1，因此，当CE3的流量到达NPE1时，由 于NPE1和UPE之间的链路失效，而导致反向流量被丢弃。为了解决该问题，业界提出了MacWithdraw功能。以图1的组网场景为例，当UPE感知到和NPE1之间链路失效时，由UPE向NPE2发送一个MacWithdraw消息，NPE2收到MacWithdraw消息后，继续向Full-Mesh域内的所有Hub PW转发该消息，同时，收到该消息后，会清除VPLS中之前所有学习到的MAC地址，除了从收到MacWithdraw消息的PW上学习到的MAC地址。对于上面这种场景，NPE3收到MacWithdraw消息后，仅保留之前从Hub PW23上学习到的MAC地址，而删除所有其他的MAC地址。这样，当CE3的流量到达NPE3后，NPE3的MAC地址表中，就不会再走之前失效的到NPE1的路径，从而改走Hub PW23到达NPE2后，再发送到CE1。从而CE1和CE3之间的转发链路保持了一致，避免的丢包产生。
RFC4762中的6.2小节定了Martini方式创建的VPLS业务中，MacWithdraw功能的实现标准。即通过对LDP的Address Withdraw消息进行扩展，增加一个Mac List TLV来携带需要撤销的MAC地址信息，当Mac List TLV中的Mac列表为空时，则表示删除之前学习到的所有MAC地址(除了从收到MacWithdraw消息的PW上学习到的MAC地址)，否则只删除Mac List TLV中指定的MAC地址。
RFC4761中没有对Kompella方式创建的VPLS业务MacWithdraw功能的实现方式进行定义。而实际上，Kompella方式创建的VPLS业务同Martini方式创建的VPLS一样也是有MAC地址加速收敛的需求的，否则当VPLS转发链路发生变化时，必然会导致反向数据流量的丢失，造成断流。
发明内容
为解决现有存在的技术问题，本发明主要提供一种MAC地址收敛方法和装置。
本发明的技术方案是这样实现的：
本发明提供一种媒体接入控制MAC地址收敛方法，该方法包括：
面向网络的提供商设备NPE间建立Kompella虚拟专用局域网服务VPLS业务，当NPE设备获知VPLS业务转发链路发生变化时，通过边界网关协议BGP消息向VPLS域中的其他NPE设备发送MAC地址撤销MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
上述方案中，所述NPE设备获知VPLS业务转发链路发生变化包括：NPE设备通过本地VPLS接入接口AC的状态变化、和/或收到接入伪线Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从备Standby变为主Active的MacWithdraw事件获知VPLS业务转发链路发生变化。
上述方案中，所述通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件包括：NPE设备通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件。
上述方案中，所述扩展BGP消息为扩展BGP更新Update消息，包括：对BGP Update消息新定义一个BGP扩展团体属性，用于携带MacWithdraw事件；
所述新定义一个BGP消息为新定义一个BGP指示Indicate消息，所述BGP Indicate消息携带MacWithdraw事件。
上述方案中，所述MacWithdraw事件携带一个以上MAC地址，或者，不携带任何有效MAC地址。
上述方案中，所述其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作，包括：其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件携带一个以上MAC地址，采用直接在本地删除 MacWithdraw事件所携带的MAC地址的策略；根据MacWithdraw事件不携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
本发明提供一种MAC地址收敛装置，该装置包括：业务建立模块、获知模块、发送模块；其中，
业务建立模块，配置为在NPE设备间建立Kompella VPLS业务；
获知模块，配置为获知VPLS业务转发链路发生变化，并通知发送模块；
发送模块，配置为通过BGP消息向VPLS域中除自身NPE设备外的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
上述方案中，所述获知模块，还配置为通过本地VPLS AC接口的状态变化、和/或收到Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从Standby变为Active的MacWithdraw事件获知VPLS业务转发链路发生变化。
上述方案中，所述发送模块，还配置为通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件。
上述方案中，该装置还包括删除模块，配置为接收到其他NPE设备发送的携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件携带一个以上MAC地址，采用直接在本地删除MacWithdraw事件所携带的MAC地址的策略；根据MacWithdraw事件不携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
本发明提供一种NPE设备，该设备包括上述的MAC地址收敛装置。
一种存储介质，所述存储介质中存储有计算机程序，所述计算机程序配置为执行前述的媒体接入控制MAC地址收敛方法。
本发明提供了一种MAC地址收敛方法和装置，NPE设备间建立Kompella VPLS业务，当NPE设备获知VPLS业务转发链路发生变化时，通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作；如此，能够在Kompella VPLS中，当VPLS转发链路发生变化时，加快各NPE设备上的MAC地址的收敛，尽量保证VPLS反向流量不丢包。
附图说明
图1为现有的VPLS网络的组网示意图；
图2为本发明实施例提供的一种MAC地址收敛方法的流程示意图；
图3为本发明实施例提供的一种MAC地址收敛装置的结构示意图；
图4为本发明实施例一提供的VPLS网络的组网示意图；
图5为本发明实施例二提供的VPLS网络的组网示意图；
图6为本发明实施例三提供的VPLS网络的组网示意图。
具体实施方式
本发明实施例中，NPE设备间建立Kompella VPLS业务，当NPE设备获知VPLS业务转发链路发生变化时，通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
下面通过附图及具体实施例对本发明做进一步的详细说明。
本发明实施例实现一种MAC地址收敛方法，如图2所示，该方法包括以下几个步骤：
步骤201：NPE设备间建立Kompella VPLS业务；
本步骤是基于RFC4761的信令流程在NPE设备间建立Kompella VPLS业务。
步骤202：NPE设备获知VPLS业务转发链路发生变化；
具体的，NPE设备可以通过本地VPLS接入接口(AC，Attachment Circuit)的状态变化、和/或收到Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从备(Standby)变为主(Active)的MacWithdraw事件获知VPLS业务转发链路发生变化。
步骤203：NPE设备通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作；
具体的，NPE设备可以通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件，所述扩展BGP消息可以是扩展BGP更新(Update)消息，如：对BGP Update消息新定义一个BGP扩展团体属性(BGP Extended Communities Attribute)，用于携带MacWithdraw事件，所述新定义一个BGP消息可以是新定义一个BGP消息，如新定义一个BGP指示(Indicate)消息，所述BGP Indicate消息携带MacWithdraw事件；
所述MacWithdraw事件可以携带一个或者多个MAC地址，也可以不携带任何有效MAC地址；
其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件携带一个或者多个MAC地址，采用直接在本地删除MacWithdraw事件所携带的MAC地址的策略，根据MacWithdraw事件不 携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
为了实现上述方法，本发明还提供一种MAC地址收敛装置，该装置设置于NPE设备中，如图3所示，该装置包括：业务建立模块31、获知模块32、发送模块33；其中，
业务建立模块31，可以由NPE设备的接口实现，配置为在NPE设备间建立Kompella VPLS业务；
获知模块32，可以由NPE设备的检测芯片实现，配置为获知VPLS业务转发链路发生变化，并通知发送模块33；
发送模块33，可以由NPE设备的处理器结合接口实现，配置为通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
所述获知模块32可以通过本地VPLS AC接口的状态变化、和/或收到Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从Standby变为Active的MacWithdraw事件获知VPLS业务转发链路发生变化。
所述发送模块33可以通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件，所述扩展BGP消息可以是扩展BGP更新(Update)消息，如：对BGP Update消息新定义一个BGP扩展团体属性，用于携带MacWithdraw事件，所述新定义一个BGP消息可以是新定义一个BGP消息，如新定义一个BGP Indicate消息，所述BGP Indicate消息携带MacWithdraw事件；
所述MacWithdraw事件可以携带一个或者多个MAC地址，也可以不携带任何有效MAC地址；
该装置还包括删除模块34，配置为接收到其他NPE设备发送的携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件携带一个或者多个MAC地址，采用直接在本地删除MacWithdraw事件所携带的MAC地址的策略，根据MacWithdraw事件不携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
基于上述装置，本发明还提供一种NPE设备，该设备包括上述图3所示的MAC地址收敛装置。
下面通过具体的场景来说明本发明方法的具体过程。
实施例一：
本实施例的场景为H-VPLS网络中双归主NPE失效，如图4所示，在H-VPLS组网中，UPE通过Spoke PW双归到Full-Mesh域中的NPE1和NPE2设备，UPE1和NPE1、NPE2之间部署的是Martini VPLS，NPE1、NPE2、NPE3、NPE4之间部署的是Kompella VPLS，稳定状态下，CE1到CE3的流量的转发路径为CE1->UPE->NPE1->NPE3->CE3；CE3到CE1的流量的转发路径为CE3->NPE3->NPE1->UPE->CE1。
当NPE1发生失效时，UPE通过PW双向转发检测(BFD)失效(Down)或者LDP会话超时等事件感知到NPE1失效，UPE立即将流量切换到Standby Spoke PW上，NPE1根据流量切换获知VPLS业务转发链路发生变化，通过扩展的BGP Update消息向NPE2发送一个空媒体接入控制列表(MacList)的MacWithdraw消息；
当NPE2收到NPE1发送来的空MacList的MacWithdraw消息后，删除VPLS域中本地之前学习到的除了从Standby Spoke PW上学习到的所有MAC地址，之后通过扩展的BGP Update消息向Full-Mesh域中的其他NPE设备，即NPE3和NPE4(NPE1已经失效，所以无法发送)继续发送空MacList 的MacWithdraw消息；
NPE3收到NPE2发送来的空MacList的MacWithdraw消息后，删除VPLS域中本地之前学习到的除了从Hub PW23上学习到的所有MAC地址；
NPE4收到NPE2发送来的空MacList的MacWithdraw消息后，删除VPLS域中本地之前学习到的除了从Hub PW24上学习到的所有MAC地址；
CE3发送给CE1的流量，到达NPE3后，由于原来的MAC地址已经被清空，CE3直接通过广播方式，把流量发送给NPE2设备，NPE2设备把流量转发给UPE，继而转发到CE1。保证了CE3到CE1的反向流量路径和CE1到CE3的流量路径为同一条路径。
实施例二：
本实施例的场景为H-VPLS网络中双归Active Spoke PW失效，如图5所示，在H-VPLS组网中，UPE通过Spoke PW双归到Full-Mesh域中的NPE1和NPE2设备，UPE1和NPE1、NPE2之间部署的是Martini VPLS，NPE1、NPE2、NPE3、NPE4之间部署的是Kompella VPLS，稳定状态下，CE1到CE3的流量的转发路径为CE1->UPE->NPE1->NPE3->CE3；CE3到CE1的流量的转发路径为CE3->NPE3->NPE1->UPE->CE1。
当Active Spoke PW发生失效时，UPE和NPE1均通过PW检测技术或者信令直接获知到VPLS业务转发链路发生变化，其中，UPE立即将流量切换到Standby Spoke PW上，NPE1通过扩展的BGP Update消息直接向Full-Mesh VPLS域中的其它所有NPE设备发送携带MacList的MacWithdraw消息，MacList中包含所有失效的MAC地址，即之前从Active Spoke PW上学习到的MAC地址；
NPE2收到NPE1发送来的带MacList的MacWithdraw消息后，把MacList中携带的MAC地址从本地VPLS转发表中删除；
NPE3收到NPE1发送来的带MacList的MacWithdraw消息后，把 MacList中携带的MAC地址从本地VPLS转发表中删除；
NPE4收到NPE1发送来的带MacList的MacWithdraw消息后，把MacList中携带的MAC地址从本地VPLS转发表中删除；
CE3发送给CE1的流量，到达NPE3后，由于原来的MAC地址已经被清空，CE3直接通过广播方式，把流量发送给NPE2设备，NPE2设备把流量转发给UPE，继而转发到CE1。保证了CE3到CE1的反向流量路径和CE1到CE3的流量路径为同一条路径。
实施例三：
本实施例的场景为Full-Mesh VPLS网络中双归Active AC失效，如图6所示，在Full-Mesh组网中，CE直接通过AC双归到Full-Mesh域中的NPE1和NPE2设备，NPE1、NPE2、NPE3、NPE4之间部署的是Kompella VPLS，稳定状态下，CE1到CE2的流量的转发路径为CE1->NPE1->NPE3->CE2；CE3到CE1的流量的转发路径为CE2->NPE3->NPE1->CE1。
当Active AC发生失效时，CE1和NPE1均通过链路检测技术直接获知到VPLS业务转发链路发生变化，其中，CE1立即将流量切换到Standby AC上，NPE1通过扩展的BGP Update消息直接向Full-Mesh VPLS域中的其它所有NPE设备发送携带MacList的MacWithdraw消息，MacList中包含所有失效的MAC地址(即之前从Active AC上学习到的MAC地址)；
NPE2收到NPE1发送来的带MacList的MacWithdraw消息后，把MacList中携带的MAC地址从本地VPLS转发表中删除；
NPE3收到NPE1发送来的带MacList的MacWithdraw消息后，把MacList中携带的MAC地址从本地VPLS转发表中删除；
NPE4收到NPE1发送来的带MacList的MacWithdraw消息后，把MacList中携带的MAC地址从本地VPLS转发表中删除；
CE2发送给CE1的流量，到达NPE3后，由于原来的MAC地址已经 被清空，CE2直接通过广播方式，把流量发送给NPE2设备，NPE2设备把流量转发给CE1。保证了CE2到CE1的反向流量路径和CE1到CE2的流量路径为同一条路径。
综上所述，通过本发明实施例提供的方案，能够在Kompella VPLS中，当VPLS转发链路发生变化时，加快各NPE设备上的MAC地址的收敛，尽量保证VPLS反向流量不丢包。
本发明实施例还记载了一种存储介质，所述存储介质中存储有计算机程序，所述计算机程序配置为执行前述各实施例的媒体接入控制MAC地址收敛方法。
以上所述，仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。
工业实用性
本发明中，NPE设备间建立Kompella VPLS业务，当NPE设备获知VPLS业务转发链路发生变化时，通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作；如此，能够在Kompella VPLS中，当VPLS转发链路发生变化时，加快各NPE设备上的MAC地址的收敛，尽量保证VPLS反向流量不丢包。

Claims (12)

  1. 一种媒体接入控制MAC地址收敛方法，该方法包括：
    面向网络的提供商设备NPE间建立Kompella虚拟专用局域网服务VPLS业务，当NPE设备获知VPLS业务转发链路发生变化时，通过边界网关协议BGP消息向VPLS域中的其他NPE设备发送MAC地址撤销MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
  2. 根据权利要求1所述的方法，其特征在于，所述NPE设备获知VPLS业务转发链路发生变化包括：NPE设备通过本地VPLS接入接口AC的状态变化、和/或收到接入伪线Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从备Standby变为主Active的MacWithdraw事件获知VPLS业务转发链路发生变化。
  3. 根据权利要求1所述的方法，其特征在于，所述通过BGP消息向VPLS域中的其他NPE设备发送MacWithdraw事件包括：NPE设备通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件。
  4. 根据权利要求3所述的方法，其特征在于，所述扩展BGP消息为扩展BGP更新Update消息，包括：对BGP Update消息新定义一个BGP扩展团体属性，用于携带MacWithdraw事件；
    所述新定义一个BGP消息为新定义一个BGP指示Indicate消息，所述BGP Indicate消息携带MacWithdraw事件。
  5. 根据权利要求4所述的方法，其特征在于，所述MacWithdraw事件携带一个以上MAC地址，或者，不携带任何有效MAC地址。
  6. 根据权利要求5所述的方法，其特征在于，所述其它NPE设备接收 到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作，包括：其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件携带一个以上MAC地址，采用直接在本地删除MacWithdraw事件所携带的MAC地址的策略；根据MacWithdraw事件不携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
  7. 一种MAC地址收敛装置，该装置包括：业务建立模块、获知模块、发送模块；其中，
    业务建立模块，配置为在NPE设备间建立Kompella VPLS业务；
    获知模块，配置为获知VPLS业务转发链路发生变化，并通知发送模块；
    发送模块，配置为通过BGP消息向VPLS域中除自身NPE设备外的其他NPE设备发送MacWithdraw事件，以使其它NPE设备接收到携带MacWithdraw事件的BGP消息后，根据MacWithdraw事件的策略执行MAC地址删除操作。
  8. 根据权利要求7所述的装置，其特征在于，所述获知模块，还配置为通过本地VPLS AC接口的状态变化、和/或收到Spoke PW状态变化、和/或接收到从Spoke PW发送过来的表明Spoke PW状态从Standby变为Active的MacWithdraw事件获知VPLS业务转发链路发生变化。
  9. 根据权利要求7所述的装置，其特征在于，所述发送模块，还配置为通过扩展BGP消息或新定义一个BGP消息利用PW向VPLS域中的其他NPE设备发送MacWithdraw事件。
  10. 根据权利要求7所述的装置，其特征在于，该装置还包括删除模块，配置为接收到其他NPE设备发送的携带MacWithdraw事件的BGP消 息后，根据MacWithdraw事件携带一个以上MAC地址，采用直接在本地删除MacWithdraw事件所携带的MAC地址的策略；根据MacWithdraw事件不携带任何有效MAC地址，采用删除本地之前学习到的除了从收到所述MacWithdraw事件的PW上学习到的所有MAC地址的策略。
  11. 一种NPE设备，其特征在于，该设备包括权利要求7至10任一项所述的MAC地址收敛装置。
  12. 一种存储介质，所述存储介质中存储有计算机程序，所述计算机程序配置为执行权利要求1至6任一项所述的媒体接入控制MAC地址收敛方法。
PCT/CN2015/077685 2014-11-25 2015-04-28 一种媒体接入控制地址收敛方法和装置、存储介质 WO2016082457A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410692185.7 2014-11-25
CN201410692185.7A CN105703996A (zh) 2014-11-25 2014-11-25 一种媒体接入控制地址收敛方法和装置

Publications (1)

Publication Number Publication Date
WO2016082457A1 true WO2016082457A1 (zh) 2016-06-02

Family

ID=56073484

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/077685 WO2016082457A1 (zh) 2014-11-25 2015-04-28 一种媒体接入控制地址收敛方法和装置、存储介质

Country Status (2)

Country Link
CN (1) CN105703996A (zh)
WO (1) WO2016082457A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113497985B (zh) * 2020-04-01 2024-04-05 上海诺基亚贝尔股份有限公司 一种上报用户的mac更新的方法与装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102377627A (zh) * 2010-08-05 2012-03-14 中兴通讯股份有限公司 一种拓扑收敛方法和服务提供商边缘设备
US8170033B1 (en) * 2009-04-06 2012-05-01 Juniper Networks, Inc. Virtual private local area network service (VPLS) flush mechanism for BGP-based VPLS networks
US20130151445A1 (en) * 2007-12-19 2013-06-13 At&T Intellectual Property I, L.P. Method and System for Survival of Data Plane Through a Total Control Plane Failure

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101621550B (zh) * 2009-08-14 2012-12-12 杭州华三通信技术有限公司 Vpls中mac地址回收方法及装置
CN101820397B (zh) * 2010-05-25 2012-10-03 杭州华三通信技术有限公司 虚拟专用局域网网络收敛的方法及设备
CN102857399B (zh) * 2012-08-20 2015-06-17 华为技术有限公司 Vpls双归属网络的mac地址删除方法、设备及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130151445A1 (en) * 2007-12-19 2013-06-13 At&T Intellectual Property I, L.P. Method and System for Survival of Data Plane Through a Total Control Plane Failure
US8170033B1 (en) * 2009-04-06 2012-05-01 Juniper Networks, Inc. Virtual private local area network service (VPLS) flush mechanism for BGP-based VPLS networks
CN102377627A (zh) * 2010-08-05 2012-03-14 中兴通讯股份有限公司 一种拓扑收敛方法和服务提供商边缘设备

Also Published As

Publication number Publication date
CN105703996A (zh) 2016-06-22

Similar Documents

Publication Publication Date Title
US20140219135A1 (en) Virtual Private Network Implementation Method and System Based on Traffic Engineering Tunnel
WO2015000375A1 (zh) 报文转发方法、装置及系统
CN103490969B (zh) 实现vpws冗余保护快速收敛的系统及方法
WO2012028029A1 (zh) 一种切换方法及系统
US20130272114A1 (en) Pseudo wire switching method and device
CN101217457A (zh) 实现快速重路由的方法及路由器
WO2008080312A1 (fr) Procédé de convergence, dispositif, système pour réseau toute connexion ainsi qu&#39;un dispositif domestique double
CN101753453A (zh) 一种分组传送网环网的组网方法
WO2011110029A1 (zh) 一种分组传送网络接入环中主机路由可达的方法及系统
CN102857418B (zh) 一种基于vpn的快速重路由切换方法及设备
WO2012075731A1 (zh) 基于arp交互的链路故障检测与恢复的方法和设备
CN102316014B (zh) 分层虚拟专用局域网服务环境下的负载分担方法及路由器
CN104811393A (zh) 组播报文复制处理方法、装置及开放流控制器
WO2012071935A1 (zh) 一种l2tp网络的保护方法及系统
WO2014019348A1 (zh) 操作、管理和维护oam配置的方法、设备及系统
WO2011147299A1 (zh) 二层虚拟专用网络快速重路由方法及装置
WO2016034127A1 (zh) 一种实现双节点互联伪线的系统及方法
WO2008011818A1 (fr) Procédé de fourniture d&#39;un service réseau local privé virtuel à hiérarchie et système réseau
WO2013139159A1 (zh) 在网络中转发报文的方法和运营商边缘设备
CN102185712B (zh) Vpls网络和以太环网的倒换方法及装置
JP6911794B2 (ja) ネットワーク装置、ネットワーク制御方法及びプログラム
CN100382501C (zh) 通信网络中三层vpn的操作维护系统及方法
WO2009100662A1 (zh) 环网的业务倒换方法和装置
CN101123563A (zh) 一种用于多跳伪线下平稳重启的方法、装置及网络
CA2619468A1 (en) Method, system and device for implementing traffic engineering

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: 15862441

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: 15862441

Country of ref document: EP

Kind code of ref document: A1