技术领域technical field
本发明涉及通信领域，具体涉及一种实现链路跟踪的方法和系统。The present invention relates to the field of communications, in particular to a method and system for implementing link tracking.
背景技术Background technique
在当前的以太网中，标准组织电气和电子工程师协会(Institute ofElectricaland Electronics Engineers，IEEE)制定了一套以太网操作、管理和维护(Operation，Administration and Maintenance，OAM)的工作机制，并已作为802.1ag标准正式发布。In the current Ethernet, the Institute of Electrical and Electronics Engineers (IEEE), a standard organization, has formulated a set of working mechanisms for Ethernet operation, management and maintenance (OAM), which has been adopted as 802.1 The ag standard is officially released.
IEEE 802.1ag标准定义了以太网中的OAM功能逻辑实体，包括位于以太网传输路径两端的维护端点(Maintenance End Point，MEP)，以及位于以太网传输路径中间的维护中间点(Maintenance Intermediate Point，MIP)。其中，维护端点和维护中间点统称为维护点(Maintenance Point，MP)。The IEEE 802.1ag standard defines the OAM functional logical entities in Ethernet, including the Maintenance End Point (MEP) located at both ends of the Ethernet transmission path, and the Maintenance Intermediate Point (MIP) located in the middle of the Ethernet transmission path. ). The maintenance endpoint and the maintenance intermediate point are collectively referred to as a maintenance point (Maintenance Point, MP).
以太网中的每台交换机(Switch，SW)都可以基于端口创建MP。连通性检测(Continuity Check，CC)和链路跟踪(LinkTrace，LT)分别是IEEE 802.1ag标准定义的以太网OAM功能之一，这两个功能可以同时运行。其中，连通性检测功能运行于MEPs之间，采用两个MEPs之间周期性主动互发连通性检测消息(Continuity Check Message，CCM)的机制，用于实时检测整个以太网传输路径的故障。链路跟踪功能运行于MEPs之间或MEP与MIP之间，当连通性检测功能检测到整个以太网传输路径没有故障时，链路跟踪功能用于进行路径发现，以检测整个或一段以太网传输路径所经过的节点情况；当连通性检测功能检测到传输路径发生故障时，链路跟踪功能用于进行故障定位。Each switch (Switch, SW) in the Ethernet can create MPs based on ports. Continuity Check (CC) and LinkTrace (LT) are respectively one of the Ethernet OAM functions defined by the IEEE 802.1ag standard, and these two functions can run simultaneously. Among them, the connectivity detection function runs between MEPs, and adopts the mechanism of periodically actively sending connectivity check messages (Continuity Check Message, CCM) between two MEPs to detect the fault of the entire Ethernet transmission path in real time. The link tracking function runs between MEPs or between MEPs and MIPs. When the connectivity detection function detects that the entire Ethernet transmission path is not faulty, the link tracking function is used for path discovery to detect the entire Ethernet transmission path or a section of the Ethernet transmission path. The status of the nodes passed through; when the connectivity detection function detects that the transmission path is faulty, the link tracking function is used to locate the fault.
参见图1，图1描述了现有IEEE 802.1ag标准定义的链路跟踪机制。图1中，SW1、SW2、SW3和SW4为四个交换机，其中SW1和SW4各创建有1个MEP，SW2和SW3各创建有1个MIP，现有以太网链路跟踪机制包括以下主要步骤：Referring to FIG. 1, FIG. 1 describes the link tracking mechanism defined by the existing IEEE 802.1ag standard. In Figure 1, SW1, SW2, SW3, and SW4 are four switches. SW1 and SW4 each create one MEP, and SW2 and SW3 each create one MIP. The existing Ethernet link tracking mechanism includes the following main steps:
步骤1，启动链路跟踪功能后，源MEP向目的MP(可以是某个MIP或对端MEP)发送链路跟踪消息(LinkTrace Message，LTM)；如图1中所示，SW1的MEP1向SW3的MIP2发送LTM。Step 1: After enabling the link tracing function, the source MEP sends a link tracing message (LinkTrace Message, LTM) to the destination MP (which may be a certain MIP or a peer MEP); as shown in FIG. 1 , MEP1 of SW1 sends a link to SW3 MIP2 sends LTM.
步骤2，源MEP到目的MP的固定单路径上的各MIP接收到所述LTM后，向发送该链路跟踪消息的源MEP回复链路跟踪响应(LinkTrace Reply，LTR)消息，该LTR消息沿反向固定单路径到达源MEP；同时，MIP会继续将所述LTM向目的MP转发；如图1中所示，MIP1向MEP1回复LTR1，同时MIP1继续将LTM向MIP2转发。Step 2: After receiving the LTM, each MIP on the fixed single path from the source MEP to the destination MP returns a link trace response (LinkTrace Reply, LTR) message to the source MEP that sent the link trace message. The reverse fixed single path reaches the source MEP; at the same time, the MIP will continue to forward the LTM to the destination MP; as shown in FIG. 1 , MIP1 replies LTR1 to MEP1, and MIP1 continues to forward the LTM to MIP2.
步骤3，目的MP接收到LTM后，向发送该LTM的源MEP回复LTR消息，该LTR消息沿反向固定单路径到达源MEP；同时，目的MP不再继续对LTM进行转发；如图1中所示，目的MP(MIP2)向MEP1回复LTR2，同时MIP2不会继续转发LTM。Step 3, after receiving the LTM, the destination MP replies an LTR message to the source MEP that sent the LTM, and the LTR message reaches the source MEP along the reverse fixed single path; at the same time, the destination MP does not continue to forward the LTM; as shown in Figure 1 As shown, the destination MP (MIP2) replies LTR2 to MEP1, and MIP2 will not continue to forward LTM.
步骤4，如果源MEP接收到所有MIP和目的MP回复的LTR消息，就达到了路径发现的目的；如果源MEP只接收到部分MIP回复的LTR消息，或者没有接收到任何LTR消息，就达到了故障定位的目的。Step 4, if the source MEP receives the LTR messages replied by all MIPs and the destination MP, the purpose of path discovery is achieved; if the source MEP only receives the LTR messages replied by part of the MIPs, or does not receive any LTR messages, it is achieved. The purpose of fault location.
IEEE 802.1Qbp项目正在制定以太网中等价多路径(Equal Cost MultiplePaths，ECMP)的工作机制，用于实现业务流量在等价多路径上的负载分担，以提高以太网中的链路利用率。IEEE 802.1Qbp标准草案(版本D0.1)规定，对于一个支持ECMP的以太网，网络边缘交换机要给进入网络的客户以太帧打上一个转发标签(F-tag)，此后网络边缘交换机和网络中间交换机根据该F-tag选择等价多路径中的一条进行客户以太帧的转发，当打上F-tag的客户以太帧离开网络时，网络边缘交换机会剥离客户以太帧所携带的F-tag。The IEEE 802.1Qbp project is developing a working mechanism of Equal Cost Multiple Paths (ECMP) in Ethernet, which is used to realize load sharing of service traffic on Equal Cost Multiple Paths, so as to improve link utilization in Ethernet. The IEEE 802.1Qbp draft standard (version D0.1) stipulates that for an Ethernet supporting ECMP, the network edge switch should add a forwarding label (F-tag) to the customer Ethernet frame entering the network, and then the network edge switch and network intermediate switch. According to the F-tag, one of the equal-cost multi-paths is selected to forward the client Ethernet frame. When the client Ethernet frame with the F-tag leaves the network, the network edge switch will strip the F-tag carried by the client Ethernet frame.
参见表1，表1描述了IEEE 802.1Qbp标准草案(版本D0.1)规定的转发标签(F-tag)的具体封装格式。Referring to Table 1, Table 1 describes the specific encapsulation format of the forwarding tag (F-tag) specified in the IEEE 802.1Qbp standard draft (version D0.1).
表1Table 1
如表1所示，F-tag包含16比特(bits)的标签协议标识(Tag Protocol ID，TPID)、3比特的优先级码组(Priority Code Point，PCP)、1比特的丢弃标识(DropEligibleIndicator，DEI)、6比特的保留字段(Reserved Field，Rev)、6比特的存活时间(Time ToLive，TTL)，以及16比特的流标识(Flow ID)。其中流标识用于对进入网络的业务流量进行分流，不同的流标识可能对应于不同的传输路径，每一个特定的流标识用于选中源节点和目的节点之间等价多路径当中的一条特定路径。而且，对于两个节点之间的双向传输路径，分别对应于两个方向特定传输路径的流标识是没有关联的。也就是说，两个节点之间的一条共路双向传输路径(所谓共路是指正向传输路径和反向传输路径所经过的节点完全一致)，对应于正向传输路径的流标识和对应于反向传输路径的流标识既可能相同，也可能不同；同样地，两个节点之间的一条不共路双向传输路径，对应于正向传输路径的流标识和对应于反向传输路径的流标识既可能相同，也可能不同。As shown in Table 1, the F-tag includes a 16-bit (bits) tag protocol identifier (Tag Protocol ID, TPID), a 3-bit priority code group (Priority Code Point, PCP), and a 1-bit drop identifier (DropEligibleIndicator, DEI), a 6-bit Reserved Field (Reserved Field, Rev), a 6-bit Time To Live (Time ToLive, TTL), and a 16-bit Flow ID (Flow ID). The flow ID is used to divide the service traffic entering the network. Different flow IDs may correspond to different transmission paths. Each specific flow ID is used to select a specific one of the equal-cost multipaths between the source node and the destination node. path. Furthermore, for a bidirectional transmission path between two nodes, the flow identities corresponding to the specific transmission paths in the two directions respectively are not associated. That is to say, a common bidirectional transmission path between two nodes (the so-called common path means that the nodes passed by the forward transmission path and the reverse transmission path are exactly the same), the flow identifier corresponding to the forward transmission path and the corresponding The flow IDs of the reverse transmission path may be the same or different; similarly, for a non-common bidirectional transmission path between two nodes, the flow ID corresponding to the forward transmission path and the flow corresponding to the reverse transmission path The identities may or may not be the same.
目前已发布的IEEE 802.1ag标准定义的链路跟踪功能，只适用两个维护点之间只有一条以太网传输路径的情况，并且该功能一般是与连通性检测功能同时使用的。要实现链路跟踪的功能，有一个前提条件，就是要求LTR所经过的返回传输路径是畅通的，这个前提条件在802.1ag标准所针对的双向共路固定单路径环境里是容易满足的。当连通性检测功能检测到传输路径没有故障时，可以确保LTR所经过的返回传输路径是畅通的；当连通性检测功能检测到传输路径发生故障时，有两种情况，一种情况是只检测到正向传输路径发生单向故障，此时同样可以确保LTR所经过的返回传输路径是畅通的，另一种情况是检测到正反向传输路径发生双向故障，此时802.1ag标准定义的链路跟踪功能只考虑该双向故障发生在传输路径同一处的情境，而不考虑两个方向的传输路径分别在不同处同时发生故障的情境，前一种情境下同样可以确保LTR所经过的返回传输路径是畅通的。The link tracking function defined by the published IEEE 802.1ag standard is only applicable to the situation where there is only one Ethernet transmission path between two maintenance points, and this function is generally used at the same time as the connectivity detection function. To realize the function of link tracking, there is a precondition that the return transmission path passed by the LTR is required to be unobstructed. This precondition is easy to satisfy in the bidirectional common-path fixed single-path environment targeted by the 802.1ag standard. When the connectivity detection function detects that the transmission path is not faulty, it can ensure that the return transmission path passed by the LTR is unobstructed; when the connectivity detection function detects that the transmission path is faulty, there are two situations, one is that only the detection When a unidirectional failure occurs in the forward transmission path, it can also ensure that the return transmission path passed by the LTR is unobstructed. Another situation is that a bidirectional failure occurs in the forward and reverse transmission paths. The path tracking function only considers the situation where the bidirectional failure occurs at the same transmission path, and does not consider the situation where the transmission paths in both directions simultaneously fail at different places. The former situation can also ensure the return transmission through which the LTR passes. The path is clear.
在IEEE 802.1Qbp所定义的ECMP以太网环境里，要求以太网OAM支持基于流的连通性检测功能和基于流的链路跟踪功能。其中，基于流的连通性检测功能可以沿用802.1ag所规定的连通性检测机制，即在两个MEPs分别发送的CCM中携带相同的Flow ID；而如果基于流的链路跟踪功能沿用802.1ag所规定的链路跟踪机制，即LTM所经过的每个MIP只回复一个与LTM携带相同Flow ID的LTR，那么很难确保LTR所经过的返回传输路径是畅通的，链路跟踪功能的目的也就难以达到。In the ECMP Ethernet environment defined by IEEE 802.1Qbp, the Ethernet OAM is required to support the flow-based connectivity detection function and the flow-based link tracking function. Among them, the flow-based connectivity detection function can follow the connectivity detection mechanism specified by 802.1ag, that is, the CCMs sent by two MEPs respectively carry the same Flow ID; The specified link tracking mechanism, that is, each MIP passed by the LTM only replies to an LTR carrying the same Flow ID as the LTM, so it is difficult to ensure that the return transmission path passed by the LTR is unobstructed, and the purpose of the link tracking function is difficult to achieve.
举例来说，如图2所示，ECMP以太网环境里有SW1、SW2、SW3等共11个交换机节点，SW1上的MEP1和SW5上的MEP2之间运行基于流的连通性检测功能，该功能通过两个MEPs之间周期性发送基于流的CCM来实现。如前所述，携带相同Flow ID的两个方向的CCM所经过的传输路径有可能不同，本例中MEP1向MEP2方向的CCM(Flow ID X)所经过的路径为SW1->SW21->SW31->SW41->SW5，MEP2向MEP1方向的CCM(Flow ID X)所经过的路径为SW5->SW42->SW33->SW24->SW1。同样地，携带相同Flow ID的LTM和LTR所经过的传输路径也有可能不同，本例中LTM(Flow ID X)所经过的路径为SW1->SW21->SW31->SW41->SW5，节点SW21上的MIP1返回的LTR1(Flow ID X)所经过的路径为SW21->SW1，节点SW31上的MIP2返回的LTR2(Flow IDX)所经过的路径为SW31->SW22->SW1，节点SW41上的MIP3返回的LTR3(Flow ID X)所经过的路径为SW41->SW32->SW23->SW1，所以本例中从MEP2向MEP1方向的CCM无法检测各个MIP返回的LTR所经过的返回传输路径是否畅通。For example, as shown in Figure 2, there are 11 switch nodes in the ECMP Ethernet environment, including SW1, SW2, SW3, etc. A flow-based connectivity detection function runs between MEP1 on SW1 and MEP2 on SW5. This is achieved by periodically sending stream-based CCMs between two MEPs. As mentioned above, the transmission paths traversed by the CCM in the two directions carrying the same Flow ID may be different. In this example, the path traversed by the CCM (Flow ID X) in the direction from MEP1 to MEP2 is SW1->SW21->SW31 ->SW41->SW5, the path taken by MEP2 to the CCM (Flow ID X) in the direction of MEP1 is SW5->SW42->SW33->SW24->SW1. Similarly, the transmission paths traversed by LTM and LTR carrying the same Flow ID may also be different. In this example, the path traversed by LTM (Flow ID X) is SW1->SW21->SW31->SW41->SW5, node SW21 The path traversed by LTR1 (Flow ID X) returned by MIP1 on the node is SW21->SW1, the path traversed by LTR2 (Flow IDX) returned by MIP2 on node SW31 is SW31->SW22->SW1, and the path on node SW41 The path traversed by LTR3 (Flow ID X) returned by MIP3 is SW41->SW32->SW23->SW1, so in this example, the CCM in the direction from MEP2 to MEP1 cannot detect whether the return transmission path traversed by the LTR returned by each MIP passes through unblocked.
可见，在ECMP以太网环境里，沿用现有技术中的链路跟踪机制完成基于流的链路跟踪功能时，很难确保LTR所经过的返回传输路径是畅通的，导致链路跟踪功能的目的难以达到。It can be seen that in the ECMP Ethernet environment, when the link tracking mechanism in the prior art is used to complete the flow-based link tracking function, it is difficult to ensure that the return transmission path traversed by the LTR is unobstructed, which leads to the purpose of the link tracking function. difficult to achieve.
发明内容SUMMARY OF THE INVENTION
有鉴于此，本发明的主要目的在于提供一种实现链路跟踪的方法和系统，尽量避免LTR所经过的返回传输路径不畅通所带来的影响，以实现ECMP场景下针对正向传输路径的链路跟踪功能。In view of this, the main purpose of the present invention is to provide a method and system for realizing link tracking, try to avoid the impact of the unsmooth return transmission path passed by the LTR, so as to realize the forward transmission path in the ECMP scenario. Link tracking function.
为达到上述目的，本发明的技术方案是这样实现的：In order to achieve the above object, the technical scheme of the present invention is achieved in this way:
一种实现链路跟踪的方法，该方法包括：A method for implementing link tracking, the method comprising:
源维护端点向目的维护点发送链路跟踪消息，其中携带所跟踪的正向传输路径的流标识；The source maintenance endpoint sends a link tracking message to the destination maintenance point, which carries the flow identifier of the tracked forward transmission path;
收到所述链路跟踪消息的维护中间点，根据自身到源维护端点的等价返回传输路径数，向源维护端点回复链路跟踪响应消息；各链路跟踪响应消息携带有不同的返回传输路径的流标识；The maintenance intermediate point that receives the link tracking message returns a link tracking response message to the source maintenance endpoint according to the number of equivalent return transmission paths from itself to the source maintenance endpoint; each link tracking response message carries a different return transmission The flow identifier of the path;
当收到的所述维护中间点回复的链路跟踪响应消息为多个时，源维护端点只对该维护中间点回复的一个链路跟踪响应消息进行处理。When there are multiple link tracking response messages replied by the maintenance intermediate point, the source maintenance endpoint only processes one link tracking response message replied by the maintenance intermediate point.
根据所述等价返回传输路径数回复链路跟踪响应消息的方法为：The method of replying to the link tracking response message according to the equivalent return transmission path number is as follows:
针对去往源维护端点的所有等价返回传输路径，维护中间点回复能够避免所经过的返回传输路径不畅通所带来的影响的多于一个的链路跟踪响应消息。For all equal-cost return transmission paths to the source maintenance endpoint, the maintenance intermediate point replies with more than one link tracking response message that can avoid the influence of the unobstructed return transmission path.
根据所述等价返回传输路径数回复链路跟踪响应消息的方法为：The method of replying to the link tracking response message according to the equivalent return transmission path number is as follows:
针对去往源维护端点的所有等价返回传输路径，维护中间点针对各返回传输路径均回复链路跟踪响应消息以覆盖所有返回传输路径；或，For all equivalent return transmission paths to the source maintenance endpoint, the maintenance intermediate point replies with a link trace response message for each return transmission path to cover all return transmission paths; or,
针对去往源维护端点的所有等价返回传输路径，维护中间点针对其中至少两条返回传输路径回复链路跟踪响应消息。For all equivalent return transmission paths to the source maintenance endpoint, the maintenance intermediate point replies with a link trace response message for at least two of the return transmission paths.
所述目的维护点为目的维护端点或目的维护中间点。The destination maintenance point is a destination maintenance endpoint or a destination maintenance intermediate point.
该方法还包括：The method also includes:
当确认所述链路跟踪响应消息所经过的反向传输路径没有故障时，目的维护端点只回复一个链路跟踪响应消息，其中携带相应的流标识；和/或，When confirming that the reverse transmission path traversed by the link tracking response message is not faulty, the destination maintenance endpoint only replies with a link tracking response message, which carries the corresponding flow identifier; and/or,
当确认所述链路跟踪响应消息所经过的反向传输路径有故障时，目的维护端点针对去往源维护端点的所有等价返回传输路径，回复能够避免所经过的返回传输路径不畅通所带来的影响的多于一个的链路跟踪响应消息。When it is confirmed that the reverse transmission path traversed by the link tracking response message is faulty, the destination maintenance endpoint responds to all the equivalent return transmission paths to the source maintenance endpoint, and the reply can avoid the unsmooth return transmission path. More than one Link Tracking Response message has been affected.
所述目的维护端点回复链路跟踪响应消息的方法为：The method for the destination maintenance endpoint to reply to the link tracking response message is:
针对去往源维护端点的所有等价返回传输路径，目的维护端点针对各返回传输路径均回复链路跟踪响应消息以覆盖所有返回传输路径；或，For all equivalent return transmission paths to the source maintenance endpoint, the destination maintenance endpoint replies with a link trace response message for each return transmission path to cover all return transmission paths; or,
针对去往源维护端点的所有等价返回传输路径，目的维护端点针对其中至少两条返回传输路径回复链路跟踪响应消息。For all equivalent return transmission paths to the source maintenance endpoint, the destination maintenance endpoint replies with a link trace response message for at least two of the return transmission paths.
该方法还包括：The method also includes:
源维护端点直接丢弃来自所述维护中间点的其它链路跟踪响应消息。The source maintenance endpoint directly discards other link tracking response messages from the maintenance intermediate point.
对所述维护中间点回复的一个链路跟踪响应消息进行处理的方法为：The method for processing a link tracking response message replied by the maintenance intermediate point is:
对所述维护中间点回复的第一个链路跟踪响应消息进行处理。Process the first link tracking response message replied by the maintenance intermediate point.
一种实现链路跟踪的系统，该系统包括源维护端点、维护中间点；其中，A system for realizing link tracking, the system includes a source maintenance endpoint and a maintenance intermediate point; wherein,
收到所述源维护端点发送的链路跟踪消息的所述维护中间点，用于根据自身到源维护端点的等价返回传输路径数，向源维护端点回复链路跟踪响应消息；各链路跟踪响应消息携带有不同的返回传输路径的流标识；The maintenance intermediate point that receives the link tracking message sent by the source maintenance endpoint is used to return the number of transmission paths according to the equivalent value from itself to the source maintenance endpoint, and reply a link tracking response message to the source maintenance endpoint; each link The trace response message carries flow identifiers of different return transmission paths;
所述源维护端点，用于向目的维护点发送链路跟踪消息，其中携带所跟踪的正向传输路径的流标识；以及在收到的所述维护中间点回复的链路跟踪响应消息为多个时，只对该维护中间点回复的一个链路跟踪响应消息进行处理。The source maintenance endpoint is configured to send a link tracking message to the destination maintenance point, which carries the flow identifier of the tracked forward transmission path; and the link tracking response message returned by the received maintenance intermediate point is multiple. When there is one, only one link tracking response message replied by the maintenance intermediate point is processed.
所述维护中间点在根据所述等价返回传输路径数回复链路跟踪响应消息时，用于：When replying a link tracking response message according to the equivalent return transmission path number, the maintenance intermediate point is used for:
针对去往源维护端点的所有等价返回传输路径，回复能够避免所经过的返回传输路径不畅通所带来的影响的多于一个的链路跟踪响应消息。For all the equivalent return transmission paths to the source maintenance endpoint, reply more than one link tracking response message that can avoid the influence of the unobstructed return transmission path.
所述维护中间点在根据所述等价返回传输路径数回复链路跟踪响应消息时，用于：When replying a link tracking response message according to the equivalent return transmission path number, the maintenance intermediate point is used for:
针对去往源维护端点的所有等价返回传输路径，针对各返回传输路径均回复链路跟踪响应消息以覆盖所有返回传输路径；或，For all equivalent return transmission paths to the source maintenance endpoint, reply a link trace response message for each return transmission path to cover all return transmission paths; or,
针对去往源维护端点的所有等价返回传输路径，针对其中至少两条返回传输路径回复链路跟踪响应消息。For all equivalent return transmission paths to the source maintenance endpoint, a link trace response message is replied for at least two of the return transmission paths.
所述目的维护点为目的维护端点或目的维护中间点。The destination maintenance point is a destination maintenance endpoint or a destination maintenance intermediate point.
当所述目的维护点为目的维护端点时，所述目的维护端点用于：When the destination maintenance point is the destination maintenance endpoint, the destination maintenance endpoint is used for:
当确认所述链路跟踪响应消息所经过的反向传输路径没有故障时，只回复一个链路跟踪响应消息，其中携带相应的流标识；和/或，When it is confirmed that the reverse transmission path traversed by the link tracking response message is not faulty, only one link tracking response message is returned, which carries the corresponding flow identifier; and/or,
当确认所述链路跟踪响应消息所经过的反向传输路径有故障时，针对去往源维护端点的所有等价返回传输路径，回复能够避免所经过的返回传输路径不畅通所带来的影响的多于一个的链路跟踪响应消息。When it is confirmed that the reverse transmission path traversed by the link tracking response message is faulty, for all the equivalent return transmission paths to the source maintenance endpoint, the reply can avoid the influence of the unsmooth return transmission path passed through. more than one link trace response message.
所述目的维护端点回复链路跟踪响应消息时，用于：When the destination maintenance endpoint replies with a link tracking response message, it is used for:
针对去往源维护端点的所有等价返回传输路径，针对各返回传输路径均回复链路跟踪响应消息以覆盖所有返回传输路径；或，For all equivalent return transmission paths to the source maintenance endpoint, reply a link trace response message for each return transmission path to cover all return transmission paths; or,
针对去往源维护端点的所有等价返回传输路径，针对其中至少两条返回传输路径回复链路跟踪响应消息。For all equivalent return transmission paths to the source maintenance endpoint, a link trace response message is replied for at least two of the return transmission paths.
所述源维护端点还用于：直接丢弃来自所述维护中间点的其它链路跟踪响应消息。The source maintenance endpoint is further configured to: directly discard other link tracking response messages from the maintenance intermediate point.
所述源维护端点在对所述维护中间点回复的一个链路跟踪响应消息进行处理时，用于：When the source maintenance endpoint processes a link tracking response message replied by the maintenance intermediate point, it is used for:
对所述维护中间点回复的第一个链路跟踪响应消息进行处理。Process the first link tracking response message replied by the maintenance intermediate point.
该系统设置于支持等价多路径的以太网中。The system is set up in an Ethernet that supports equal-cost multipathing.
可见，收到源维护端点发送的链路跟踪消息的维护中间点，能够根据自身到源维护端点的等价返回传输路径数，向源维护端点回复链路跟踪响应LTR消息；各链路跟踪响应消息携带有不同的返回传输路径的流标识；当收到的所述维护中间点回复的链路跟踪响应消息为多个时，源维护端点只对该维护中间点回复的一个链路跟踪响应消息进行处理。本发明方法和系统，均可尽量避免LTR所经过的返回传输路径不畅通所带来的影响，以实现ECMP场景下针对正向传输路径的链路跟踪功能。It can be seen that the maintenance intermediate point that receives the link tracking message sent by the source maintenance endpoint can return the number of transmission paths according to the equivalent value between itself and the source maintenance endpoint, and reply a link tracking response LTR message to the source maintenance endpoint; each link tracking response The messages carry flow identifiers of different return transmission paths; when there are multiple link tracking response messages replied by the maintenance intermediate point, the source maintenance endpoint only returns one link tracking response message to the maintenance intermediate point to be processed. The method and system of the present invention can avoid the influence brought by the unsmooth return transmission path passed by the LTR as much as possible, so as to realize the link tracking function for the forward transmission path in the ECMP scenario.
附图说明Description of drawings
图1为IEEE 802.1ag标准规定的链路跟踪过程示意图；Figure 1 is a schematic diagram of the link tracking process specified by the IEEE 802.1ag standard;
图2为ECMP环境中沿用802.1ag标准规定的链路跟踪机制示意图；Figure 2 is a schematic diagram of the link tracking mechanism specified by the 802.1ag standard in the ECMP environment;
图3为本发明基于流的链路跟踪流程简图；Fig. 3 is the flow-based link tracing flow chart of the present invention;
图4为本发明实施例的用于路径验证的链路跟踪过程示意图；4 is a schematic diagram of a link tracking process for path verification according to an embodiment of the present invention;
图5为本发明实施例的用于故障定位的链路跟踪过程示意图。FIG. 5 is a schematic diagram of a link tracking process for fault location according to an embodiment of the present invention.
具体实施方式Detailed ways
在实际应用中，在支持等价多路径的以太网，通过扩展IEEE 802.1ag标准定义的链路跟踪机制，基于流的传输路径上的维护中间点可以根据与源维护端点之间的等价路径数，向源维护端点回复能够覆盖所有等价返回传输路径的链路跟踪响应消息。源维护端点则只处理从所述维护中间点返回的一个链路跟踪响应消息(如第一个链路跟踪响应消息)，因而能够尽量避免LTR所经过的返回传输路径不畅通所带来的影响。In practical applications, in the Ethernet that supports equal-cost multi-path, by extending the link tracking mechanism defined by the IEEE 802.1ag standard, the maintenance intermediate point on the flow-based transmission path can be based on the equivalent path between the source maintenance endpoint number, and reply to the source maintenance endpoint with a link trace response message that can cover all equivalent return transmission paths. The source maintenance endpoint only processes one link tracking response message (such as the first link tracking response message) returned from the maintenance intermediate point, so it can try to avoid the impact of the unsmooth return transmission path passed by the LTR as much as possible. .
参见图3，图3为本发明基于流的链路跟踪流程简图，该流程包括以下步骤：Referring to FIG. 3, FIG. 3 is a schematic diagram of the flow-based link tracking process flow of the present invention, and the process flow includes the following steps:
步骤301：源维护端点向目的维护点发送链路跟踪消息。Step 301: The source maintenance endpoint sends a link tracking message to the destination maintenance point.
所述目的维护点可能是对端维护端点(可称为目的维护端点)，也可能是某个维护中间点(可称为目的维护中间点)。The destination maintenance point may be a peer maintenance endpoint (which may be referred to as a destination maintenance endpoint), or may be a certain maintenance intermediate point (which may be referred to as a destination maintenance intermediate point).
所述链路跟踪消息携带有F-tag，其中包含所跟踪的特定正向传输路径的流标识(通常为16比特)。The link tracking message carries an F-tag, which contains the flow identification (usually 16 bits) of the specific forward transmission path being tracked.
步骤302：维护中间点收到所述链路跟踪消息后，向源维护端点回复一个或多个链路跟踪响应消息。需要说明的是，回复的链路跟踪响应消息的个数等于所述维护中间点到源维护端点的等价返回传输路径数，且所有链路跟踪响应消息分别携带不同的流标识并遵循不同的返回传输路径。然后，所述维护中间点向目的维护点继续转发所述链路跟踪消息。Step 302: After receiving the link tracking message, the maintenance intermediate point returns one or more link tracking response messages to the source maintenance endpoint. It should be noted that the number of reply link tracking response messages is equal to the number of equivalent return transmission paths from the maintenance intermediate point to the source maintenance endpoint, and all link tracking response messages carry different flow identifiers and follow different flow identifiers. Return the transmission path. Then, the maintenance intermediate point continues to forward the link tracking message to the destination maintenance point.
IEEE 802.1Qbp标准草案(版本D0.1)规定，支持ECMP的以太网运行中间系统到中间系统(Intermediate System to Intermediate System，IS-IS)路由协议，所以每一个网络节点都学习到了全网拓扑；而且，利用流标识和本地节点信息从等价多路径中选择特定传输路径的哈希(Hash)算法是标准化的，所以维护中间点可以自动选择一个或多个不同的流标识，进而分别封装一个或多个不同的链路跟踪响应消息，以实现对所有等价返回传输路径的全覆盖。The IEEE 802.1Qbp draft standard (version D0.1) stipulates that the Ethernet supporting ECMP runs the Intermediate System to Intermediate System (IS-IS) routing protocol, so each network node learns the entire network topology; Moreover, the Hash algorithm for selecting a specific transmission path from the equal-cost multipath using the flow ID and local node information is standardized, so the maintenance intermediate point can automatically select one or more different flow IDs, and then encapsulate one or multiple different link tracking response messages to achieve full coverage of all equivalent return transmission paths.
步骤303：源维护端点收到所述维护中间点回复的链路跟踪响应消息后，记录下其中携带的节点标识、入接口、出接口等节点信息。并且，如果收到的所述维护中间点回复的链路跟踪响应消息(即：携带所述维护中间点的节点标识的链路跟踪响应消息)为多个，则源维护端点只对这些消息中的一个链路跟踪响应消息(如第一个链路跟踪响应消息)进行处理，而直接丢弃来自所述维护中间点的其它链路跟踪响应消息。Step 303: After receiving the link tracking response message replied by the maintenance intermediate point, the source maintenance endpoint records the node information such as the node identifier, the inbound interface, the outbound interface and the like carried in the message. In addition, if there are multiple link tracking response messages (that is, link tracking response messages carrying the node identifier of the maintenance intermediate point) replied by the maintenance intermediate point, the source maintenance endpoint only responds to the link tracking response messages in these messages. One of the link tracking response messages (eg, the first link tracking response message) is processed, and other link tracking response messages from the maintenance intermediate point are directly discarded.
对于启动链路跟踪功能的源维护端点而言，该源维护端点只关心所跟踪的特定正向传输路径所经过节点的可达性及其节点信息，而并不关心这些节点的返回传输路径信息，也就是说并不关心链路跟踪响应消息是通过什么路径返回的。所以针对所跟踪的特定正向传输路径经过的每个MIP，只要有一个链路跟踪响应消息能够返回到源维护端点，链路跟踪功能的实现就不会受到影响。For the source maintenance endpoint that starts the link tracking function, the source maintenance endpoint only cares about the reachability and node information of the nodes passed by the specific forward transmission path being tracked, but does not care about the return transmission path information of these nodes. , that is to say, it does not care about the path through which the link tracking response message is returned. Therefore, for each MIP that the tracked specific forward transmission path passes through, as long as there is a link tracking response message that can be returned to the source maintenance endpoint, the implementation of the link tracking function will not be affected.
以下通过两个实际应用中的具体实施例对本发明进一步描述。The present invention is further described below through two specific embodiments in practical applications.
实施例一、Embodiment 1.
本实施例中，源维护端点与对端维护端点之间运行基于流的连通性检测功能，周期性互发携带Flow ID X的CCM消息，通过运行连通性检测功能判断出MEP1到MEP2的对应于Flow ID X的特定正向传输路径没有故障。然后，源维护端点向对端维护端点发起基于流的链路跟踪，用于进行路径验证，以检测从源维护端点向对端维护端点的以太网传输路径所经过的节点情况。这里之所以称为路径验证，而不是802.1ag标准中所称的路径发现，是因为802.1Qbp标准草案(版本D0.1)规定，在支持ECMP的以太网中要运行IS-IS路由协议，网络中的每一个节点都通过该路由协议学习到了全网的拓扑和所有的传输路径，所以链路跟踪在这里所实现的是路径验证的功能，即验证实际的传输路径是否与通过IS-IS路由协议学习到的传输路径一致。In this embodiment, a flow-based connectivity detection function is run between the source maintenance endpoint and the peer maintenance endpoint, and CCM messages carrying Flow ID X are periodically exchanged. The specific forward transmission path of Flow ID X is not faulty. Then, the source maintenance endpoint initiates flow-based link tracking to the peer maintenance endpoint for path verification, so as to detect the situation of the nodes passed by the Ethernet transmission path from the source maintenance endpoint to the peer maintenance endpoint. The reason why it is called path verification here, rather than path discovery in the 802.1ag standard, is because the 802.1Qbp standard draft (version D0.1) stipulates that the IS-IS routing protocol should be run in the Ethernet that supports ECMP, and the network Each node has learned the topology of the entire network and all transmission paths through this routing protocol, so link tracking here is the function of path verification, that is, to verify whether the actual transmission path is the same as the route through IS-IS. The transmission paths learned by the protocol are consistent.
参见图4，图4为本发明实施例的用于路径验证的链路跟踪过程示意图。结合图3和图4，则图4所示过程主要包括如下步骤：Referring to FIG. 4 , FIG. 4 is a schematic diagram of a link tracking process for path verification according to an embodiment of the present invention. Combined with Figure 3 and Figure 4, the process shown in Figure 4 mainly includes the following steps:
步骤401：源维护端点MEP1根据输入信息确定目的维护点为对端维护端点MEP2并向其发送LTM，该LTM中携带所跟踪的传输路径的Flow ID X。Step 401: The source maintenance endpoint MEP1 determines the destination maintenance point as the peer maintenance endpoint MEP2 according to the input information and sends an LTM to it, where the LTM carries the Flow ID X of the tracked transmission path.
源维护端点接收的输入信息包括MEP2所在端口的MAC地址，还包括从MEP1到达MEP2的特定正向传输路径的标识，这个标识可以是Flow ID X，也可以是能够计算出FlowID X的其它标识，比如描述流的信息参数(参数集“源/目的MAC地址、源/目的IP地址、源/目的端口”中的一个或多个)。The input information received by the source maintenance endpoint includes the MAC address of the port where MEP2 is located, and also includes the identifier of the specific forward transmission path from MEP1 to MEP2. This identifier can be Flow ID X or other identifiers that can calculate FlowID X. For example, information parameters describing the flow (one or more of the parameter set "source/destination MAC address, source/destination IP address, source/destination port").
所跟踪的传输路径的Flow ID X包含在LTM所携带的F-tag中，于是LTM会遵循所述传输路径到达MEP2。The Flow ID X of the tracked transmission path is included in the F-tag carried by the LTM, so the LTM will follow the transmission path to reach MEP2.
步骤402：维护中间点MIP1收到LTM后，通过判断发现只有一条去往源维护端点MEP1的返回传输路径SW21->SW1。于是MIP1只向MEP1回复一个链路跟踪响应消息LTR1，其中携带从所述LTM中拷贝的Flow ID X。然后MIP1继续向目的维护点MEP2转发所述LTM。Step 402: After the maintenance intermediate point MIP1 receives the LTM, it is found through judgment that there is only one return transmission path SW21->SW1 to the source maintenance endpoint MEP1. Therefore, MIP1 only replies to MEP1 a link tracking response message LTR1, which carries the Flow ID X copied from the LTM. Then MIP1 continues to forward the LTM to the destination maintenance point MEP2.
维护中间点MIP2收到LTM后，通过判断发现有两条去往源维护端点MEP1的返回传输路径SW31->SW21->SW1和SW31->SW22->SW1，且Flow ID X和Flow ID Y分别对应于这两条返回传输路径。于是MIP2向MEP1回复两个链路跟踪响应消息LTR21和LTR22以覆盖两条返回传输路径，LTR21和LTR22分别携带Flow ID X和Flow ID Y。然后MIP2继续向目的维护点MEP2转发所述LTM。After the maintenance intermediate point MIP2 receives the LTM, it is found through judgment that there are two return transmission paths SW31->SW21->SW1 and SW31->SW22->SW1 to the source maintenance endpoint MEP1, and Flow ID X and Flow ID Y are respectively Corresponding to these two return transmission paths. Then MIP2 replies to MEP1 with two link tracking response messages LTR21 and LTR22 to cover the two return transmission paths, LTR21 and LTR22 carrying Flow ID X and Flow ID Y respectively. Then MIP2 continues to forward the LTM to the destination maintenance point MEP2.
维护中间点MIP3收到LTM后，通过判断发现有三条去往源维护端点MEP1的返回传输路径SW41->SW31->SW21->SW1、SW41->SW31->SW22->SW1和SW41->SW32->SW23->SW1，且Flow ID X、Flow ID Y和Flow ID Z分别对应于这三条返回传输路径。于是MIP3向MEP1回复三个链路跟踪响应消息LTR31、LTR32和LTR33以覆盖三条返回传输路径，LTR31、LTR32和LTR33分别携带Flow ID X、Flow ID Y和Flow ID Z。然后MIP3继续向目的维护点MEP2转发所述LTM。After the maintenance intermediate point MIP3 receives the LTM, it is determined that there are three return transmission paths SW41->SW31->SW21->SW1, SW41->SW31->SW22->SW1 and SW41->SW32 to the source maintenance endpoint MEP1 ->SW23->SW1, and Flow ID X, Flow ID Y, and Flow ID Z correspond to the three return transmission paths, respectively. Then MIP3 replies three link tracking response messages LTR31, LTR32 and LTR33 to MEP1 to cover the three return transmission paths, LTR31, LTR32 and LTR33 carry Flow ID X, Flow ID Y and Flow ID Z respectively. Then MIP3 continues to forward the LTM to the destination maintenance point MEP2.
需要说明的是，根据接收到的CCM消息中的远端缺陷指示(RemoteDefectIndicator，RDI)，如果通过判断发现MEP2到MEP1的对应于Flow ID X的特定反向传输路径没有故障，那么目的维护点MEP2收到LTM后，可以只向MEP1回复一个链路跟踪响应消息LTR4，其中携带Flow ID X；如果通过判断发现MEP2到MEP1的对应于Flow ID X的特定反向传输路径有故障，那么目的维护点MEP2可以根据判断出的四条去往MEP1的返回传输路径，向MEP1回复四个链路跟踪响应消息，其中分别携带可以覆盖这四条返回传输路径的不同的Flow ID。It should be noted that, according to the remote defect indication (RemoteDefectIndicator, RDI) in the received CCM message, if it is found through judgment that the specific reverse transmission path corresponding to Flow ID X from MEP2 to MEP1 is not faulty, then the destination maintenance point MEP2 After receiving the LTM, it can only reply a link tracking response message LTR4 to MEP1, which carries Flow ID X; if the specific reverse transmission path corresponding to Flow ID X from MEP2 to MEP1 is found to be faulty through judgment, the destination maintenance point MEP2 may reply four link tracking response messages to MEP1 according to the determined four return transmission paths to MEP1, which respectively carry different Flow IDs that can cover the four return transmission paths.
步骤403：源维护端点MEP1收到维护中间点MIP1、MIP2、MIP3，以及目的维护点MEP2回复的链路跟踪响应消息后，记录下其中携带的节点标识、入接口、出接口等节点信息。如果上述的所有返回传输路径都是畅通的，源维护端点MEP1会收到MIP2回复的两个链路跟踪响应消息，以及MIP3回复的三个链路跟踪响应消息，但MEP1只对分别来自MIP2和MIP3的第一个链路跟踪响应消息进行处理，而直接丢弃来自MIP2和MIP3的其它链路跟踪响应消息。Step 403: After receiving the link tracking response message from the maintenance intermediate points MIP1, MIP2, MIP3, and the destination maintenance point MEP2, the source maintenance endpoint MEP1 records the node information such as the node identifier, inbound interface, and outbound interface carried therein. If all the above-mentioned return transmission paths are unobstructed, the source maintenance endpoint MEP1 will receive two link tracking response messages replied by MIP2 and three link tracking response messages replied by MIP3, but MEP1 only responds to messages from MIP2 and MIP3 respectively. The first link tracking response message of MIP3 is processed, and other link tracking response messages from MIP2 and MIP3 are directly discarded.
实施例二、Embodiment two,
本实施例中，源维护端点与对端维护端点之间运行基于流的连通性检测功能，周期性互发携带Flow ID X的CCM消息，通过运行连通性检测功能判断出MEP1到MEP2的对应于Flow ID X的特定正向传输路径有故障。然后，源维护端点向对端维护端点发起基于流的链路跟踪，用于进行故障定位，以确定从源维护端点向对端维护端点的以太网传输路径上是哪两个节点之间存在故障。In this embodiment, a flow-based connectivity detection function is run between the source maintenance endpoint and the peer maintenance endpoint, and CCM messages carrying Flow ID X are periodically exchanged. The specific forward transmission path of Flow ID X is faulty. Then, the source maintenance endpoint initiates flow-based link tracking to the peer maintenance endpoint for fault location to determine which two nodes have a fault on the Ethernet transmission path from the source maintenance endpoint to the peer maintenance endpoint .
参见图5，图5为本发明实施例的用于故障定位的链路跟踪过程示意图。结合图3和图5，则图5所示过程主要包括如下步骤：Referring to FIG. 5 , FIG. 5 is a schematic diagram of a link tracking process for fault location according to an embodiment of the present invention. Combined with Figure 3 and Figure 5, the process shown in Figure 5 mainly includes the following steps:
步骤501：源维护端点MEP1根据输入信息确定目的维护点为对端维护端点MEP2并向其发送LTM，该LTM中携带所跟踪的传输路径的Flow ID X。Step 501: The source maintenance endpoint MEP1 determines the destination maintenance point as the peer maintenance endpoint MEP2 according to the input information, and sends an LTM to it, where the LTM carries the Flow ID X of the tracked transmission path.
源维护端点接收的输入信息包括MEP2所在端口的MAC地址，还包括从MEP1到达MEP2的特定正向传输路径的标识，这个标识可以是Flow ID X，也可以是能够计算出FlowID X的其它标识，比如描述流的信息参数(参数集“源/目的MAC地址、源/目的IP地址、源/目的端口”中的一个或多个)。The input information received by the source maintenance endpoint includes the MAC address of the port where MEP2 is located, and also includes the identifier of the specific forward transmission path from MEP1 to MEP2. This identifier can be Flow ID X or other identifiers that can calculate FlowID X. For example, information parameters describing the flow (one or more of the parameter set "source/destination MAC address, source/destination IP address, source/destination port").
所跟踪的传输路径的Flow ID X包含在LTM所携带的F-tag中。The Flow ID X of the traced transmission path is contained in the F-tag carried by the LTM.
步骤502：维护中间点MIP1收到LTM后，通过判断发现只有一条去往源维护端点MEP1的返回传输路径SW21->SW1，于是MIP1只向MEP1回复一个链路跟踪响应消息LTR1，其中携带从所述LTM中拷贝的Flow ID X。然后MIP1继续向目的维护点MEP2转发所述LTM。Step 502: After the maintenance intermediate point MIP1 receives the LTM, it is found through judgment that there is only one return transmission path SW21->SW1 to the source maintenance endpoint MEP1, so MIP1 only returns a link tracking response message LTR1 to MEP1, which carries the Flow ID X copied in the LTM described above. Then MIP1 continues to forward the LTM to the destination maintenance point MEP2.
维护中间点MIP2收到LTM后，通过判断发现有两条去往源维护端点MEP1的返回传输路径SW31->SW21->SW1和SW31->SW22->SW1，且Flow ID X和Flow ID Y分别对应于这两条返回传输路径。于是MIP2向MEP1回复两个链路跟踪响应消息LTR21和LTR22以覆盖两条返回传输路径，LTR21和LTR22分别携带Flow ID X和Flow ID Y。然后MIP2继续向目的维护点MEP2转发所述LTM。After the maintenance intermediate point MIP2 receives the LTM, it is found through judgment that there are two return transmission paths SW31->SW21->SW1 and SW31->SW22->SW1 to the source maintenance endpoint MEP1, and Flow ID X and Flow ID Y are respectively Corresponding to these two return transmission paths. Then MIP2 replies to MEP1 with two link tracking response messages LTR21 and LTR22 to cover the two return transmission paths, LTR21 and LTR22 carrying Flow ID X and Flow ID Y respectively. Then MIP2 continues to forward the LTM to the destination maintenance point MEP2.
由于从MIP2去往MIP3的对应于Flow ID X的正向传输路径存在故障，所以维护中间点MIP3无法收到所述LTM，也就不会发送任何链路跟踪回复消息。Since the forward transmission path corresponding to Flow ID X from MIP2 to MIP3 is faulty, the maintenance intermediate point MIP3 cannot receive the LTM, and thus will not send any link tracking reply message.
步骤503：源维护端点MEP1收到维护中间点MIP1和MIP2回复的链路跟踪响应消息后，记录下其中携带的节点标识、入接口、出接口等节点信息。如果上述的所有返回传输路径都是畅通的，源维护端点MEP1会收到MIP2回复的两个链路跟踪响应消息，但MEP1只对来自MIP2的第一个链路跟踪响应消息进行处理，而直接丢弃来自MIP2的另一个链路跟踪响应消息。Step 503: After receiving the link tracking response messages returned by the maintenance intermediate points MIP1 and MIP2, the source maintenance endpoint MEP1 records the node information such as the node identifier, inbound interface, and outbound interface carried in the message. If all the above-mentioned return transmission paths are unobstructed, the source maintenance endpoint MEP1 will receive two link tracking response messages replied by MIP2, but MEP1 only processes the first link tracking response message from MIP2, and directly Another link trace response message from MIP2 is discarded.
可见，通过对比本次链路跟踪所收集到的Flow ID X所对应的正向传输路径上的节点信息，以及之前收集到的Flow ID X所对应的正向传输路径上的节点信息，可以判断出传输故障发生在MIP2与MIP3之间(包括MIP3)。It can be seen that by comparing the node information on the forward transmission path corresponding to Flow ID X collected in this link tracking, and the node information on the forward transmission path corresponding to Flow ID X collected previously, it can be determined that Outgoing transmission failure occurs between MIP2 and MIP3 (including MIP3).
结合图3以及上述两个实施例可知，收到链路跟踪消息的维护中间点，可以根据自身到源维护端点的等价返回传输路径数，向源维护端点回复链路跟踪响应消息；所有链路跟踪响应消息都分别携带有相应的返回传输路径的流标识。With reference to Fig. 3 and the above two embodiments, it can be seen that the maintenance intermediate point that receives the link tracking message can return the number of transmission paths according to the equivalent value from itself to the source maintenance endpoint, and reply a link tracking response message to the source maintenance endpoint; The path trace response messages all carry the flow identifiers of the corresponding return transmission paths.
基于上述思路，为了保证可靠性，针对去往源维护端点的所有返回传输路径，维护中间点会针对各返回传输路径均回复链路跟踪响应消息以覆盖所有返回传输路径。但在实际应用中，也可以进行各种灵活变化，如：针对去往源维护端点的所有返回传输路径，维护中间点针对其中至少两条返回传输路径回复链路跟踪响应消息，或者针对其中不少于两个、并且一半以上的返回传输路径回复链路跟踪响应消息。可见，无论返回传输路径的数目如何，只要所回复的具有一定数量(尽量多于一个)的链路跟踪响应消息能够尽量避免LTR所经过的返回传输路径不畅通所带来的影响即可。上述思路同样适用于步骤402中目的维护点MEP2回复链路跟踪响应消息的情况。Based on the above ideas, in order to ensure reliability, for all return transmission paths to the source maintenance endpoint, the maintenance intermediate point will reply a link tracking response message for each return transmission path to cover all return transmission paths. However, in practical applications, various flexible changes can also be made, such as: for all return transmission paths to the source maintenance endpoint, the maintenance intermediate point replies with link tracking response messages for at least two of the return transmission paths, or for no of the return transmission paths. Less than two and more than half of the return transmission paths reply to the link trace response message. It can be seen that no matter what the number of return transmission paths is, as long as a certain number (more than one) of the link tracking response messages replied can try to avoid the influence of unsmooth return transmission paths passed by the LTR. The above idea is also applicable to the situation in which the destination maintenance point MEP2 in step 402 replies with a link tracking response message.
另外，选择要回复链路跟踪响应消息的返回传输路径时，可以考虑通信负载、传输速率、服务质量(QoS)等。In addition, the communication load, transmission rate, quality of service (QoS), etc. may be considered when selecting the return transmission path to which the link trace response message is to be replied.
综上所述可见，无论是实现链路跟踪的方法还是包含维护中间点、源维护端点的实现链路跟踪的系统，本发明均可尽量避免LTR所经过的返回传输路径不畅通所带来的影响，以实现ECMP场景下针对正向传输路径的链路跟踪功能。From the above, it can be seen that, whether it is a method for realizing link tracking or a system for realizing link tracking including maintenance intermediate points and source maintenance endpoints, the present invention can try to avoid the problems caused by the unsmooth return transmission path passed by the LTR as much as possible. In order to implement the link tracking function for the forward transmission path in the ECMP scenario.
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。The above descriptions are merely preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110419929.4ACN103166786B (en) | 2011-12-15 | 2011-12-15 | A method and system for implementing link tracking |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110419929.4ACN103166786B (en) | 2011-12-15 | 2011-12-15 | A method and system for implementing link tracking |
| Publication Number | Publication Date |
|---|---|
| CN103166786A CN103166786A (en) | 2013-06-19 |
| CN103166786Btrue CN103166786B (en) | 2019-03-08 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110419929.4AExpired - Fee RelatedCN103166786B (en) | 2011-12-15 | 2011-12-15 | A method and system for implementing link tracking |
| Country | Link |
|---|---|
| CN (1) | CN103166786B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103607301B (en)* | 2013-11-18 | 2018-04-06 | 新华三技术有限公司 | The protection switching method and equipment of link failure under a kind of ECMP modes |
| CN106452963A (en)* | 2016-10-18 | 2017-02-22 | 北京东土科技股份有限公司 | Network path sensing method and system |
| US10129127B2 (en)* | 2017-02-08 | 2018-11-13 | Nanning Fugui Precision Industrial Co., Ltd. | Software defined network controller, service function chaining system and trace tracking method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101826989A (en)* | 2009-03-02 | 2010-09-08 | 华为技术有限公司 | Method and device for treating fault |
| CN101986604A (en)* | 2010-10-29 | 2011-03-16 | 中兴通讯股份有限公司 | Link fault positioning method and system of packet transport network (PTN) |
| CN102177681A (en)* | 2011-04-21 | 2011-09-07 | 华为技术有限公司 | Method and system for detecting faults |
| CN102185719A (en)* | 2011-05-12 | 2011-09-14 | 中兴通讯股份有限公司 | Service link tracking realization method and device |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8391163B2 (en)* | 2009-03-23 | 2013-03-05 | Cisco Technology, Inc. | Operating MPLS label switched paths and MPLS pseudowire in loopback mode |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101826989A (en)* | 2009-03-02 | 2010-09-08 | 华为技术有限公司 | Method and device for treating fault |
| CN101986604A (en)* | 2010-10-29 | 2011-03-16 | 中兴通讯股份有限公司 | Link fault positioning method and system of packet transport network (PTN) |
| CN102177681A (en)* | 2011-04-21 | 2011-09-07 | 华为技术有限公司 | Method and system for detecting faults |
| CN102185719A (en)* | 2011-05-12 | 2011-09-14 | 中兴通讯股份有限公司 | Service link tracking realization method and device |
| Publication number | Publication date |
|---|---|
| CN103166786A (en) | 2013-06-19 |
| Publication | Publication Date | Title |
|---|---|---|
| CN1716912B (en) | Method and apparatus providing rapid end-to-end failover in a packet switched communications network | |
| CN104509044B (en) | Enhancing Protocol Independent Multicast (PIM) Fast Rerouting Method with Downstream Notification Packets | |
| US8792509B2 (en) | In-band signaling for point-multipoint packet protection switching | |
| CN102684940B (en) | The packet loss at the service endpoints place of Virtual Private LAN Service is measured | |
| CN103873368B (en) | Ethernet message forwarding method and access device | |
| EP2498454A1 (en) | Method, device and system for processing service traffic based on pseudo wires | |
| CN104380671B (en) | The increase failure covering in classification, redundancy, QoS routing selection | |
| CN104662851A (en) | Enhancements to PIM fast re-route with upstream activation packets | |
| CN101179479A (en) | Transmission method, system and node of Ethernet operation management and maintenance message | |
| US8902794B2 (en) | System and method for providing N-way link-state routing redundancy without peer links in a network environment | |
| US9893929B2 (en) | Protection switching method and system for a multi-rooted point-to-multi-point service in a provider backbone bridge (PBB) network | |
| CN104869010B (en) | Protection switching | |
| CN109981323B (en) | Method and network equipment for detecting multicast path state of data link layer | |
| US20110170403A1 (en) | Service Movement in Link State Controlled Layer Two Networks | |
| WO2013189414A2 (en) | Automatic network topology acquisition method and system, and network query and management system | |
| CN104348717A (en) | Message forwarding method and device | |
| US8612626B2 (en) | Group member detection among nodes of a network | |
| US8559431B2 (en) | Multiple label based processing of frames | |
| CN103166786B (en) | A method and system for implementing link tracking | |
| CN102833122B (en) | Loopback detection method and system | |
| CN103973471B (en) | A kind of notifying method and device of TRILL distribution trees failure | |
| JP2014502063A (en) | Communication path verification system, path verification apparatus, communication path verification method, and path verification program | |
| CN103634210A (en) | Method and apparatus for discovering opposite-end provider edge (PE) device of virtual private LAN service (VPLS) instance | |
| WO2020168982A1 (en) | Method for sending and obtaining assert message and network node | |
| CN103036731B (en) | A kind of equipment of the loopback detection based on stream, system and method |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20190308 |