WO2011022905A1 - 单环网络拓扑重建方法及系统 - Google Patents

单环网络拓扑重建方法及系统 Download PDF

Info

Publication number
WO2011022905A1
WO2011022905A1 PCT/CN2009/075682 CN2009075682W WO2011022905A1 WO 2011022905 A1 WO2011022905 A1 WO 2011022905A1 CN 2009075682 W CN2009075682 W CN 2009075682W WO 2011022905 A1 WO2011022905 A1 WO 2011022905A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
port
state
blocked
ring network
Prior art date
Application number
PCT/CN2009/075682
Other languages
English (en)
French (fr)
Inventor
张恺
冷卫杰
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
Priority to US13/392,199 priority Critical patent/US20120155245A1/en
Priority to EP09848637.6A priority patent/EP2472795B1/en
Publication of WO2011022905A1 publication Critical patent/WO2011022905A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/42Loop networks
    • H04L12/437Ring fault isolation or reconfiguration
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/22Alternate routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/28Routing or path finding of packets in data switching networks using route fault recovery

Definitions

  • the present invention relates to the field of communications, and in particular to a single-loop network topology reconstruction method and system.
  • STP Spanning-Tree Protocol
  • the Spanning-Tree Protocol is a two-layer protocol that discovers physical loops in a network through a dedicated algorithm and generates a logical loop-free (loop-free) Topological structure.
  • the main features of the Spanning Tree Protocol include:
  • the Spanning Tree Protocol will reconfigure the ports of the switch to avoid link loss or new loops.
  • FIG. 1 is a schematic diagram of a network topology reconstruction process based on the STP protocol in the related art.
  • the ring network topology is converted into a tree topology, that is, a form in which the first sub-picture in FIG. 1 is converted into a second sub-picture.
  • the process of network topology reconstruction begins, and the network topology structure is changed from the second sub-picture to the third sub-picture form.
  • the root bridge is the root node of the tree structure, that is, node A shown in Figure 1.
  • the root port is the port on the non-root bridge that forwards data frames to the root bridge (ie, the parent node), each non-root There is only one root port on the bridge.
  • the port of the node B of the node B shown in Figure 1 is connected to the node A (port R shown in the figure); the selected port is the root bridge or the non-bridge bridge
  • the port in which the child node forwards data for example, another port of node B (port D shown in the figure).
  • all ports of the root bridge are selected ports.
  • the role of each node and the state of the port in the node are elected.
  • the election process compares some parameters of each node through BPDUs to select the root bridge, root port, and selected port.
  • the STP protocol specifies four states of the port: blocking state, listening state ( listening ) , learning state ( learning ) and forward state (also known as forwarding data state , forwording ) , blocking is blocking state , ie, data cannot be forwarded, forwarding is the state in which data can be forwarded, and the other two states are from blocking Switching to the intermediate state of the forwarding state, STP stipulates that only the port and the port can be used to forward data.
  • the port and the selected port are given the forwarding state, and the other ports are in the blocking state.
  • the topology is The process of establishment is complete.
  • the current standard STP protocol follows the IEEE 802.1D standard. Please refer to the standard for the specific election process. In addition to these key concepts, the STP protocol specifies several key time parameters:
  • forward delay The STP protocol specifies four states of the port: blocking, listening, learning, and forwording. The forwarding delay is the duration of the listening or learning state. The default is 15s.
  • max age Maximum lifetime, the maximum valid time of BPDU messages. The default value is 20s.
  • STP uses the above time parameters to control the transmission of standard BPDU messages, state transitions, and when to perform topology reconstruction.
  • the salient problem with the standard STP protocol is that the convergence time is too long when the topology changes.
  • the topology reconstruction process is as follows: Suppose there are three bridges running STP. They are connected to the ring network as shown in Figure 2. After the election process is completed, the P1 port of bridge C becomes blocking. If the link connecting A and C fails, the P2 port of C does not receive the standard BPDU message. As a result, the standard BPDU message storage time of the P2 port exceeds max age, and bridge C re-elects the election process.
  • P1 will be elected as the designated port, so that P1 will be converted from the blocking state to the forwarding state, according to the state transition diagram shown in FIG. 3, during which time two forwarding delays are experienced.
  • the STP protocol it takes up to 50s to calculate the topology reconstruction by default.
  • the nature of the STP protocol determines whether the network topology is complicated or simple, and the determinants of the topology reconstruction speed include several time parameters rather than the election process.
  • each port on each bridge in the network re-elects its own state very quickly, and most of the time is essentially used to discover network topology changes and state transitions of ports.
  • the selection of these time parameters is related to the number of nodes in the network and the network topology. Reducing the above time parameters repeatedly may result in the network topology not being stably established, and the potential for reducing the above time parameters is also limited.
  • the main reason for the long convergence time of the standard STP protocol is: The characteristics of passive waiting. For example, in the listening phase, the bridge has learned that a previously blocked port does not generate a loop after entering the forwarding state, and cannot enter the learning state immediately. It needs to wait for 15 seconds of forwarding delay to expire and enter the learning state. If the port of the bridge learns all the MAC addresses in a short period of time during the learning phase, it cannot immediately enter the forwarding state because it has to wait for a 15 second forwarding delay.
  • the present invention has been made in view of the problem of slow convergence at the time of STP-based network topology reconstruction in the related art. Therefore, the main object of the present invention is to provide an improved network topology reconstruction method and system to solve At least one of the above issues. According to an aspect of the present invention, a single loop network topology reconstruction method is provided.
  • the single-loop network topology reconstruction method includes: each node in a single-ring network detects whether a link connected to the root port of the node is invalid according to a predetermined time interval; when a node in the single-ring network detects When the link fails, the node with the blocked port learns the link failure information in the single ring network; the node with the blocked port converts the state of the blocked port to the forwarding data state.
  • a single loop network topology reconstruction system is provided.
  • the single-loop network topology reconstruction system includes a plurality of nodes, and the plurality of nodes include a node having a blocked port, wherein the plurality of nodes are configured to detect the root port of the node according to a predetermined time interval. Whether the connected link is invalid; a node with a blocked port is used to learn the link failure information in the single ring network and detect it in the blocked state when a node in the single ring network detects the link failure. The state is converted to the forwarding data state.
  • each node in the single ring network actively detects whether the link connected to the root port of the node is invalid; when a node in the single ring network detects the link failure, the node with the blocked port obtains a single The information about the link failure occurs in the ring network.
  • the node with the blocked port converts the state of the blocked port to the forwarding data state, which solves the problem of slow convergence when the STP-based network topology reconstruction is implemented in the related art.
  • the passive waiting for the link failure can be changed to actively detect the link failure, and the state of the blocked port is directly converted into the forwarding number. According to the state, the speed of convergence is accelerated.
  • FIG. 1 is a schematic diagram of a network topology reconstruction process based on STP protocol in the related art
  • FIG. 2 is a schematic diagram of a single ring network including three nodes
  • FIG. 3 is a STP state transition diagram of a node port; A flowchart of a single-loop network topology reconstruction method according to an embodiment of the present invention; FIG. 5 is a flowchart of a single-loop network topology reconstruction method according to a preferred embodiment of the present invention; FIG. 6 is a flowchart according to an embodiment of the present invention.
  • FIG. 7 is a block diagram showing the functional structure of a node in a single ring network according to a preferred embodiment of the present invention;
  • FIG. 8 is a single ring network topology reconstruction system according to a preferred embodiment of the present invention; Schematic diagram of the structure.
  • the embodiments of the present invention provide a single-loop network topology reconstruction scheme.
  • each node has only two ports.
  • Most of the time for STP topology reconstruction is used in the discovery of network topology changes and port state transitions. Combined with the characteristics of single ring networks, the following advantages are analyzed:
  • the new configuration message will be propagated to the entire network after a certain delay. This delay is called Forward Delay, and the default value of the protocol is 15 seconds.
  • Forward Delay the default value of the protocol is 15 seconds.
  • the spanning tree uses a timer policy, and an intermediate state in which the port learns only the MAC address but does not participate in forwarding, that is, the learning state, is added from the blocked state to the forwarding state. This problem does not exist under a single loop condition, so the learning state can be skipped.
  • the single-loop network topology reconstruction method according to the embodiment of the present invention includes the following processing (step S401 - step S405): Step 4: S401: Each node in the single-ring network detects the node according to a predetermined time interval. Whether the link connected to the port is invalid or not; according to the standard STP protocol, when the link fails, the BPDU of the root port can only be passively waited for the max age to expire, and then the topology change information can be forwarded to the root bridge. propagation.
  • the technical solution provided by the embodiment of the present invention can actively detect the link state of the root port.
  • the software When the root port of a certain bridge detects a link fault (for example, detecting a handshake signal, etc.), the software implements actively A timer that controls the expiration of the BPDU—the max age timer expires, and the topology occurrence message is transmitted to the entire topology.
  • the active detection mode includes but is not limited to: detecting by a preset time, and the preset time may be set according to actual needs.
  • Step S403 When a node in the single ring network detects a link failure, the node having the blocked port learns the information about the link failure occurring in the single ring network; preferably, if it is connected to the root port of the node having the blocked port If the link fails, the node actively detects that the link connected to the port of the node is invalid.
  • the node having the non-blocking port detects that the link connected to the root port of the node is invalid; thereafter, having the non-blocking port
  • the node transmits the customized BPDU message to the next node through the re-elected selected port until reaching the node with the blocked port; the node with the blocked port receives the BPDU message and obtains the link failure information in the network.
  • the above customized BPDU message has a different type from the standard BPDU message. For example, the field of the identification type in the BPDU message can be changed.
  • Step S405 The node having the blocked port converts the state of the port in the blocked state to the forwarding data state.
  • a non-standard type BPDU message is added.
  • the defined BPDU message is used to convey the topology change message.
  • the type I of the BPDU may be different from the configuration BPDU and the TCN BPDU.
  • the non-standard BPDU is sent through the selected port of the node.
  • other nodes After receiving this type of BPDU, other nodes will also time out its max age timer and repeat the process until there is a node with a blocking status port.
  • the state of the port is switched to the forward state via the listening state. That is, the learning state is skipped, and according to the STP protocol, the 15 S time can be reduced.
  • the single-loop network topology reconstruction method in the preferred embodiment of the present invention includes the following processing (step S501 - step S519):
  • Step S501 A node in the single-ring network detects that the link is faulty, or the node receives a non-standard BPDU (Custom BPDU) message, where the node does not have a blocked port;
  • Step S503 The node causes the max age timer to expire, and then enters a max age timer timeout processing flow;
  • Step S505 Sending a non-standard type BPDU message (frame);
  • Step S507 Designating the port as the selected port;
  • Step S509 Performing the root port election again;
  • Step S511 Performing the selected port election again;
  • Step S513 The node determines whether the port has the blocking status, if If yes, step 4 is performed, S515.
  • step 4 is performed.
  • the node does not have a blocking state
  • the non-standard type BPDU message is sent through the selected port of the node, and the other node receives the message. After the type of BPDU, it will also time out its max age timer, repeat this process until there is blocking Node port.
  • Step S515 The node having the blocked port converts the state of the port in the blocked state (blocking state) into the forwarding data state (the forwarding state).
  • S517 The node with the blocked port continues to send the TCN message to the next node, where the message is used to notify the other node of the network that the link fails.
  • Step S519 The node with the non-blocking port performs port state assignment.
  • each node actively detects the link failure, and then actively propagates the topology change information to the network, and finally actively changes the port state.
  • This scheme is used on a bridge with a frequency of 200 MHz using PPC as the CPU and Linux as the operating system, and the speed of network topology reconstruction is about seconds. Therefore, it can be seen that in the process of topology reconstruction of a single ring network, the use of this scheme can effectively shorten the time of topology reconstruction, plus Fast STP convergence speed.
  • a single ring network topology reconstruction system is further provided.
  • 6 is a schematic structural diagram of a single-loop network topology reconstruction system according to an embodiment of the present invention
  • FIG. 7 is a functional structural block diagram of a node in a single-ring network according to a preferred embodiment of the present invention
  • FIG. 8 is a preferred embodiment of the present invention.
  • a single ring network topology reconstruction system includes: a plurality of nodes (the five nodes are shown as an example: node A, node B, node C, node G, node E) ), wherein the plurality of nodes include one node with a blocked port (node G), and the other nodes are nodes with non-blocking ports.
  • a plurality of nodes the five nodes are shown as an example: node A, node B, node C, node G, node E)
  • the plurality of nodes include one node with a blocked port (node G)
  • the other nodes are nodes with non-blocking ports.
  • the foregoing multiple nodes are configured to detect, according to a predetermined time interval, whether a link connected to the root port of the node is invalid; a node having a blocked port among the multiple nodes, configured to detect a node in the single ring network
  • the link fails, the information about the link failure in the single-ring network is learned, and the state of the port in the blocked state is converted into the forwarding data state.
  • the node having the blocked port in the single ring network may include: the first detecting unit 1 is configured to detect whether the link connected to the root port is invalid.
  • the node having the non-blocking port according to the function division may include: a second detecting unit 2 and a sending unit 3.
  • the second detecting unit 2 is configured to detect whether the link connected to the root port is invalid.
  • the sending unit 3 is configured to send the customized BPDU message to the next node by using the re-elected selected port.
  • the node having the blocked port in the single-ring network may further include: a receiving unit 4, configured to receive a customized BPDU message, where the customized BPDU message has non-blocking The node of the port is sent.
  • the plurality of nodes in the single-loop network topology reconstruction system are configured to: when the port state of any one of the multiple nodes changes, the state is blocked by the listening state. Directly transition to the forward state. For example, as shown in FIG.
  • a single-loop network topology reconstruction system includes a node with a blocked port and a node with a non-blocking port, and actively detects link faults through each node, and then actively The network propagates the topology change information, and finally changes the port state actively, which can greatly shorten the time of topology reconstruction.
  • the single-loop network topology reconstruction solution provided by the foregoing embodiment of the present invention actively detects link faults when actively reconstructing the network topology, and then actively propagates topology change information to the network, and finally Actively changing the port state, solving the problem of slow convergence when STP-based network topology reconstruction in the related art, and then converting the passive waiting for the link failure to actively detecting the link failure, and making the blocked state port
  • the state is directly converted to the forwarding data state, which can greatly shorten the time of topology reconstruction, thereby speeding up the convergence.
  • the implementation of the present invention does not modify the system architecture and the current processing flow, is easy to implement, facilitates promotion in the technical field, and has strong industrial applicability.
  • modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or they may be Multiple modules or steps are made into a single integrated circuit module.
  • the invention is not limited to any specific combination of hardware and software.
  • the above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Description

单环网络拓朴重建方法及系统 技术领域 本发明涉及通信领域, 具体而言, 涉及一种单环网络拓朴重建方法及系 统。 背景技术 生成树协议 (Spanning-Tree Protocol, 简称为 STP ) 是一种两层协议, 其通过一种专用的算法来发现网络中的物理环路, 并产生一个逻辑的无环 ( loop-free ) 拓朴结构。 生成树协议的主要功能包括:
( 1 ) 维持一个无环路的网络, 避免网络风暴, 如果一个设备在拓朴中 发现一个回路, 它将阻塞一个或多个冗余的端口, 使环形拓朴转变成 †形拓 朴。
( 2 ) 当网络拓朴发生变化时, 生成树协议将重新配置交换机的各个端 口以避免链接丢失或者出现新的回路。
STP实施以上两项功能的示意图可以参见图 1 , 图 1为相关技术中基于 STP协议的网络拓朴重建过程的示意图。 如图 1所示, 环状网络拓朴转换成 了树形拓朴, 即由图 1中的第一副图转换成第二副图的形式。 当节点 G和 E 之间的链路出现故障的时候, 开始网络拓朴重建的过程, 网络拓朴结构由第 二副图转变成了第三副图的形式。 为了清楚了解 STP协议, 首先介绍 STP中几个关键的概念: ( 1 )根桥、 ( 2 )根端口、 ( 3 )选取端口。 可以参见图 1 , 根桥是树形结构的根节点, 即 图 1中所示节点 A; 根端口是非根桥上向根桥方向 (即, 父节点) 转发数据 帧的端口, 每个非根桥上只有一个根端口,例如, 图 1中所示的节点 B的两个 端口中与节点 A连接链路的端口 (图中所示端口 R ); 选取端口是根桥或非 才艮桥向子节点方向转发数据的端口, 例如, 节点 B的另一端口 (图中所示端 口 D )。 其中, 根桥(节点 A ) 的所有端口都是选取端口。 在 STP协议中, 各个节点的角色和节点中端口的状态是选举产生的, 选举过程就是通过 BPDU将各个节点的一些参数进行比较,从而选出根桥,根端口和选取端口。 STP 协议中规定了端口的四种状态: 阻塞状态 ( blocking ) , 监听状态 ( listening ) , 学习状态 ( learning ) 和前向状态 (又可称转发数据状态, forwording ), blocking 是阻塞状态, 即, 不能转发数据, forwarding 是可以 转发数据的状态, 其它两个状态是从 blocking转换到 forwarding状态的中间 状态, STP规定只用才艮端口和选取端口可以转发数据, 所以将 -端口和选取 端口赋予 forwarding状态, 其它端口都是 blocking状态, 各个节点的端口状 态稳定后, 拓朴建立的过程也就完成了。 现行的标准 STP 协议遵循 IEEE 802.1D标准,具体的选举过程请参看标准。除了这几个关键的概念之外, STP 协议还规定了几个关键的时间参数:
( 1 ) hello time:发送标准网桥十办议数据单元 ( Bridge Protocol Data Unit, 简称为 BPDU ) 消息的间隔。
( 2 ) forward delay: 转发延迟, STP 协议规定了端口的四种状态: blocking, listening, learning和 forwording。 forwarding delay是 listening或 learning状态的持续时间, 缺省值为 15s。
( 3 ) max age: 最大存活周期, BPDU消息的最大有效时间, 缺省值为 20s。
STP使用上述时间参数, 控制标准 BPDU消息的发送, 状态的转换和 判断何时进行拓朴的重建。 标准 STP协议的突出问题是当拓朴发生变化时收敛时间过长。根据 STP 协议, 拓朴重建流程如下: 假设现在有三个运行 STP协议的网桥, 它们如图 2所示连接成环网, 选举过程完成后, 网桥 C的 P1端口成为 blocking状态。 若连接 A和 C的链路发生故障, 则 C的 P2端口接收不到标准 BPDU消息, 导致 P2端口的标准 BPDU消息存储时间超过 max age, 网桥 C重新进行选 举过程。此时,才艮据 STP协议 P1将被选举为指定端口,这样, P1将从 blocking 状态转换成 forwarding状态, 根据如图 3所示的状态转换图, 此间要经历两 个 forwarding delay的时间。 可见按照标准 STP协议, 按缺省值计算拓朴重 建最多需要 50s的时间。 通过对 STP协议的分析可知: STP协议的性质决定了不论网络拓朴结 构如何复杂或简单, 拓朴重建速度的决定因素包括几个时间参数, 而不是选 举过程。 实际上, 网络中各网桥上的各端口重新选举自己的状态是十分迅速 的, 大部分时间实质上用在了发现网络拓朴变化和端口的状态转换上。 而这 些时间参数的选择与网络中节点的数量和网络拓朴是有联系的, 一再减小上 述时间参数可能导致网络拓朴不能稳定建立, 同时减小上述时间参数的潜力 也是有限的。 标准 STP协议收敛时间长的主要原因是: 被动等待的特性。 例 如, 网桥在 listening阶段用了很短的时间就已经获知某原先阻塞的端口进入 转发状态后不会产生环路, 无法立即进入学习状态, 需要等待 15 秒的转发 延迟时间过期进入学习状态; 如果网桥的端口在学习阶段经过很短的时间就 学习完了所有的 MAC地址, 也无法立即进入转发状态, 因为它还要等待 15 秒的转发延迟。 发明内容 针对相关技术中基于 STP 的网络拓朴重建时, 收敛速度较慢的问题而 提出本发明, 为此, 本发明的主要目的在于提供一种改进的网络拓朴重建方 法及系统, 以解决上述问题至少之一。 才艮据本发明的一个方面, 提供了一种单环网络拓朴重建方法。 才艮据本发明的单环网络拓朴重建方法包括:单环网络中每个节点按照预 定时间间隔检测与该节点根端口相连接的链路是否失效; 当单环网络中的一 个节点检测到链路失效时, 具有阻塞端口的节点获知单环网络中发生链路失 效的信息; 具有阻塞端口的节点将其处于阻塞状态的端口的状态转换成转发 数据状态。 根据本发明的另一方面, 提供了一种单环网络拓朴重建系统。 才艮据本发明的单环网络拓朴重建系统包括多个节点,多个节点中包含一 个具有阻塞端口的节点, 其中, 上述多个节点, 用于按照预定时间间隔检测 与该节点根端口相连接的链路是否失效; 具有阻塞端口的节点, 用于在单环 网络中的一个节点检测到链路失效时,获知单环网络中发生链路失效的信息, 并将其处于阻塞状态的端口的状态转换成转发数据状态。 通过本发明,单环网络中的每个节点主动检测与该节点根端口相连接的 链路是否失效; 当该单环网络中的一个节点检测到链路失效时, 具有阻塞端 口的节点获知单环网络中发生链路失效的信息; 具有阻塞端口的节点将其处 于阻塞状态的端口的状态转换成转发数据状态, 解决了相关技术中基于 STP 的网络拓朴重建时, 收敛速度较慢的问题, 进而可以将被动等待获知链路失 效转变为主动检测链路失效, 并使阻塞状态的端口的状态直接转换成转发数 据状态, 从而加快了收敛速度。 本发明的其它特征和优点将在随后的说明书中阐述, 并且, 部分地从说 明书中变得显而易见, 或者通过实施本发明而了解。 本发明的目的和其他优 点可通过在所写的说明书、 权利要求书、 以及附图中所特别指出的结构来实 现和获得。 附图说明 此处所说明的附图用来提供对本发明的进一步理解,构成本申请的一部 分, 本发明的示意性实施例及其说明用于解释本发明, 并不构成对本发明的 不当限定。 在附图中: 图 1为相关技术中基于 STP协议的网络拓朴重建过程的示意图; 图 2为包含 3个节点的单环网络示意图; 图 3为节点端口的 STP状态转换图; 图 4为才艮据本发明实施例的单环网络拓朴重建方法的流程图; 图 5为根据本发明优选实施例的单环网络拓朴重建方法的流程图; 图 6为才艮据本发明实施例的单环网络拓朴重建系统的结构示意图; 图 7为根据本发明优选实施例的单环网络中的节点的功能结构框图; 图 8为根据本发明优选实施例的单环网络拓朴重建系统的结构示意图。 具体实施方式 功能相克述 本发明实施例提供了一种单环网络拓朴重建方案, 对于单环网, 每个节 点只具有两个端口。针对 STP拓朴重建的大部分时间实质上用在了发现网络 拓朴变化和端口的状态转换上这一特性, 结合单环网的特点分析出以下的优 4匕方向:
( 1 ) 当拓朴发生变化, 新的配置消息要经过一定的时延才能传播到整 个网络, 这个时延称为 Forward Delay, 协议默认值是 15秒。 在所有网桥收 到这个变化的消息之前, 若旧拓朴结构中处于转发的端口还没有发现自己应 该在新的拓朴中停止转发, 则可能存在临时环路。 为了解决临时环路的问题, 生成树使用了一种定时器策略, 在端口从阻塞状态到转发状态中间加上一个 只学习 MAC地址但不参与转发的中间状态, 即, learning状态。 在单一环路 条件下是不存在这个问题的, 所以可以将 learning状态跳过。
( 2 ) 链路失效后, STP需要等最多 20s的时间才可以发现链路发生变 化。 如果可以在链路出现故障时马上发现, 并强制进入拓朴变化处理流程, 则可以节省最多 20s的重建时间。
( 3 ) 由于是单环网, 系统中只有一个节点有 blocking状态的端口, 如 果出链路故障, 则 blocking端口是与它相连的节点通向根节点的唯一途径。 所以, 在检测到链路故障的时候, 直接将 blocking 状态的端口转换成 forwarding , 跳过 listening和 learning (优化后没有 learning状态了 ) 状态。 需要说明的是, 在不冲突的情况下, 本申请中的实施例及实施例中的特 征可以相互组合。 下面将参考附图并结合实施例来详细说明本发明。 方法实施例 才艮据本发明实施例, 首先提供了一种单环网络拓朴重建方法。 图 4 为才艮据本发明实施例的单环网络拓朴重建方法的流程图。 如图 4 所示,根据本发明实施例的单环网络拓朴重建方法包括以下处理(步骤 S401- 步骤 S405 ): 步 4聚 S401 : 单环网络中每个节点按照预定时间间隔检测与该节点才艮端 口相连接的链路是否失效; 根据标准 STP协议,在链路发生故障时,只能被动等待根端口的 BPDU 消息到达 max Age而过期, 此后才可以将拓朴变化信息向根桥方向传播。 优选地, 本发明实施例提供的技术方案可以主动检测根端口的链路状 态, 当某网桥的根端口检测到链路故障 (例如, 检测握手信号等) 后, 在软 件实现上主动地使控制 BPDU过期的定时器—— max age定时器过期, 将拓 朴发生消息向整个拓朴结构传送。 其中, 主动检测的方式包括但不限于: 以 预设时间进行检测, 预设时间可以根据实际需求设置。 步骤 S403:当单环网络中的一个节点检测到链路失效时, 具有阻塞端口 的节点获知单环网络中发生链路失效的信息; 优选地, 如果是与具有阻塞端口的节点的根端口相连的链路出现故障, 则该节点主动检测到与该节点 4艮端口相连接的链路失效。 优选地, 如果是与具有非阻塞端口的节点的根端口相连的链路出现故 障, 则具有非阻塞端口的节点检测到与该节点根端口相连接的链路失效; 之 后, 具有非阻塞端口的节点通过重新选举的选取端口向下一节点传送自定义 的 BPDU 消息直至到达具有阻塞端口的节点; 具有阻塞端口的节点接收到 BPDU消息, 获取网络中链路失效的信息。 其中, 上述自定义的 BPDU消息与标准的 BPDU消息具有不同的类型。 例如, 可以将 BPDU消息中标识类型的字段进行更改。 步骤 S405 : 具有阻塞端口的节点将其处于阻塞状态的端口的状态转换 成转发数据状态。 由于是单环网, 系统中只有一个有 blocking状态的端口的网桥, 如果出 现链路故障, 则 blocking端口是通向根网桥的唯一途径。 所以, 在检测到链 路故障的时候, 直接将 blocking状态的端口转换成 forwarding状态, 而其他 端口的 ^1 态不变。 当某一端口检测到拓朴变化时, 为了将拓朴变化消息快速通知到有 blocking 状态端口的网桥, 根据本发明实施例提供的技术方案, 增加了一种 非标准类型的 BPDU消息( 自定义的 BPDU消息)来传递拓朴变化消息。 此 种 BPDU的类型 i或不同于配置 BPDU和 TCN BPDU即可, 当某一个节点发 现链路故障后, 如果此节点没有 blocking状态的端口, 则通过此节点的选取 端口发送非标准类型的 BPDU, 其它节点接收到此种类型的 BPDU后, 也会 将它的 max age定时器超时, 重复这个过程, 直至有 blocking状态端口的节 点。 优选地, 如果在重建拓朴时, 需要加入新的节点, 对于单环网络中需要 增加的节点, 当该节点的端口处于阻塞状态时, 使端口的状态经由监听状态 转换至前向状态。 即跳过学习状态, 才艮据 STP协议, 可以减少 15 S时间。 图 5为根据本发明优选实施例的单环网络拓朴重建方法的流程图。如图 5 所示, 居本发明优选实施例的单环网络拓朴重建方法包括以下处理 (步 骤 S501-步骤 S519 ): 步骤 S501 : 单环网络中一个节点检测到链路出现故障, 或者该节点接 收到非标准的 BPDU ( 自定义的 BPDU ) 消息, 其中, 该节点不具有阻塞端 口; 步骤 S503 : 该节点使 max age定时器超时, 则进入 max age定时器超时 处理流程; 步骤 S505 : 发送非标准类型的 BPDU消息 (帧); 步骤 S507: 将端口指定为选取端口; 步骤 S509: 重新进行根端口选举; 步骤 S511: 重新进行选取端口选举; 步骤 S 513: 该节点判断是否具有 blocking状态的端口, 如果是, 则执 行步 4聚 S515 , 否则, 执行步 4聚 S519; 优选地, 如果该节点不具有 blocking状态的端口, 则通过此节点的选取 端口发送非标准类型的 BPDU消息, 其它节点接收到此种类型的 BPDU后, 也会将它的 max age定时器超时, 重复这个过程, 直至有 blocking状态端口 的节点。 步骤 S515 : 具有阻塞端口的节点将其处于阻塞状态的端口的状态 ( blocking状态) 转换成转发数据状态 ( forwarding状态)。 步 4聚 S517:具有阻塞端口的节点继续向下一节点发送 TCN消息,其中, 该消息用于通知网络中其他节点链路失效的信息; 步骤 S519: 具有非阻塞端口的节点进行端口状态的分配。 通过上述实施例, 在网络拓朴重建过程中, 通过各节点主动的检测链路 故障, 然后主动地向网络传播拓朴变化信息, 最后主动的更改了端口状态, 在具体实施过程中, 在主频为 200MHz的使用 PPC作为 CPU和 Linux作为 操作系统的网桥上使用该方案, 网络拓朴重建的速度在秒级左右。 因而可知 在单环网络拓朴重建过程中, 使用该方案可以有效缩短拓朴重建的时间, 加 快 STP收敛速度。 系统实施例 根据本发明实施例, 还提供了一种单环网络拓朴重建系统。 图 6为 居本发明实施例的单环网络拓朴重建系统的结构示意图; 图 7 为 居本发明优选实施例的单环网络中的节点的功能结构框图; 图 8为才艮据 本发明优选实施例的单环网络拓朴重建系统的结构示意图。 如图 6所示, 根据本发明实施例的单环网络拓朴重建系统包括: 多个节 点 (图中所示以五个节点为例: 节点 A、 节点 B、 节点 C、 节点 G、 节点 E ), 其中, 该多个节点包含一个具有阻塞端口的节点(节点 G ), 其他节点为具有 非阻塞端口的节点。 以下结合图 7、 图 8进行描述。 其中, 上述多个节点, 用于按照预定时间间隔检测与该节点根端口相连 接的链路是否失效; 多个节点中的具有阻塞端口的节点,用于在单环网络中的一个节点检测 到链路失效时, 获知单环网络中发生链路失效的信息, 并将其处于阻塞状态 的端口的状态转换成转发数据状态。 优选地, 如图 7所示, 单环网络中具有阻塞端口的节点按照功能可以包 括: 第一检测单元 1 , 用于检测与其根端口相连接的链路是否失效。 优选地, 如图 7所示, 上述具有非阻塞端口的节点按照功能划分可以包 括: 第二检测单元 2、 发送单元 3。 其中, 第二检测单元 2 , 用于检测与其根 端口相连接的链路是否失效; 发送单元 3 , 用于通过重新选举的选取端口向 下一节点传送自定义的 BPDU消息。 优选地, 如图 7所示, 单环网络中具有阻塞端口的节点按照功能划分还 可以包括: 接收单元 4 , 用于接收自定义的 BPDU消息, 其中, 该自定义的 BPDU消息由具有非阻塞端口的节点发送。 优选地, 如图 8所示, 上述单环网络拓朴重建系统中的多个节点: 用于 在所述多个节点中任一节点的端口状态发生转变时, 使状态由阻塞状态经过 监听状态直接转换至前向状态。 例如, 如图 8所示, 对于节点端口状态发生转变的一种情况, 需要在单 环网络中增加一个或多个节点(图中示出了一个: 节点 F ), 节点 F的端口状 态发生转变时, 状态由阻塞状态经过监听状态直接转换至前向状态。 通过上述实施例, 提供了一种单环网络拓朴重建系统, 该系统包括的一 个具有阻塞端口的节点, 以及具有非阻塞端口的节点, 通过各节点主动的检 测链路故障, 然后主动的向网络传播拓朴变化信息, 最后主动的更改了端口 状态, 可以大大缩短拓朴重建的时间。 综上所述, 通过本发明的上述实施例提供的单环网络拓朴重建方案, 在 对网络拓朴重建时, 首先主动地检测链路故障, 然后主动地向网络传播拓朴 变化信息, 最后主动地更改端口状态, 解决了相关技术中基于 STP的网络拓 朴重建时, 收敛速度较慢的问题, 进而可以将被动等待获知链路失效转变为 主动检测链路失效, 并使阻塞状态的端口的状态直接转换成转发数据状态, 可以大大缩短拓朴重建的时间, 从而可以加快收敛速度。 另夕卜,本发明的实现没有对系统架构和目前的处理流程修改,易于实现, 便于在技术领域中进行推广, 具有较强的工业适用性。 显然, 本领域的技术人员应该明白, 上述的本发明的各模块或各步骤可 以用通用的计算装置来实现, 它们可以集中在单个的计算装置上, 或者分布 在多个计算装置所组成的网络上, 可选地, 它们可以用计算装置可执行的程 序代码来实现, 从而, 可以将它们存储在存储装置中由计算装置来执行, 或 者将它们分别制作成各个集成电路模块, 或者将它们中的多个模块或步骤制 作成单个集成电路模块来实现。 这样, 本发明不限制于任何特定的硬件和软 件结合。 以上所述仅为本发明的优选实施例而已, 并不用于限制本发明, 对于本 领域的技术人员来说, 本发明可以有各种更改和变化。 凡在本发明的 ^"神和 原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明的保护 范围之内。

Claims

权 利 要 求 书
1. 一种单环网络拓朴重建方法, 其特征在于, 包括:
单环网络中每个节点按照预定时间间隔检测与该节点才艮端口相连 接的链路是否失效;
当所述单环网络中的一个节点检测到链路失效时,具有阻塞端口的 节点获知所述单环网络中发生链路失效的信息;
所述具有阻塞端口的节点将其处于阻塞状态的端口的状态转换成 转发数据状态。
2. 根据权利要求 1所述的方法, 其特征在于, 具有阻塞端口的节点获知所 述单环网络中发生链路失效的信息包括:
所述具有阻塞端口的节点通过检测获知与该节点根端口相连接的 链路失效。
3. 根据权利要求 1所述的方法, 其特征在于, 具有阻塞端口的节点获知所 述单环网络中发生链路失效的信息包括:
具有非阻塞端口的节点通过检测获知与该节点根端口相连接的链 路失效;
所述具有非阻塞端口的节点通过选取端口向下一节点传送自定义 的桥接协议数据单元 BPDU消息直至到达所述具有阻塞端口的节点; 所述具有阻塞端口的节点接收到所述 BPDU消息,获取网络中链路 失效的信息。
4. 根据权利要求 3所述的方法, 其特征在于, 所述自定义的 BPDU消息与 标准的 BPDU消息包含的标识类型的字段不相同。
5. 根据权利要求 1至 4中任一项所述的方法, 其特征在于, 所述方法还包 括:
对于所述单环网络中的任一节点, 当该节点的端口状态发生转变 时, 状态转变过程为由阻塞状态经由监听状态直接转换至前向状态。
6. 一种单环网络拓朴重建系统, 其中, 所述系统包括多个节点, 所述多个 节点包含一个具有阻塞端口的节点, 其特征在于,
所述多个节点,用于按照预定时间间隔检测与该节点根端口相连接 的链路是否失效;
所述具有阻塞端口的节点,用于在所述单环网络中的一个节点检测 到链路失效时, 获知所述单环网络中发生链路失效的信息, 并将其处于 阻塞状态的端口的状态转换成转发数据状态。
7. 根据权利要求 6所述的系统, 其特征在于, 所述具有阻塞端口的节点包 括:
第一检测单元, 用于检测与其根端口相连接的链路是否失效。
8. 根据权利要求 6所述的系统, 其特征在于,
所述多个节点中具有非阻塞状态的节点包括:
第二检测单元, 用于检测与其根端口相连接的链路是否失效; 发送单元,用于通过重新选举的选取端口向下一节点传送自定义的 桥接协议数据单元 BPDU消息;
所述具有阻塞端口的节点还包括:
接收单元, 用于接收所述自定义的 BPDU消息。
9. 根据权利要求 6至 8中任一项所述的系统, 其特征在于,
所述多个节点,用于在所述多个节点中任一节点的端口状态发生转 变时, 使状态由阻塞状态经过监听状态直接转换至前向状态。
PCT/CN2009/075682 2009-08-25 2009-12-17 单环网络拓扑重建方法及系统 WO2011022905A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/392,199 US20120155245A1 (en) 2009-08-25 2009-12-17 Method and System for Rebuilding Single Ring Network Topology
EP09848637.6A EP2472795B1 (en) 2009-08-25 2009-12-17 Method and system for re-building single ring network topology

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200910167562.4 2009-08-25
CN2009101675624A CN101997735A (zh) 2009-08-25 2009-08-25 单环网络拓扑重建方法及系统

Publications (1)

Publication Number Publication Date
WO2011022905A1 true WO2011022905A1 (zh) 2011-03-03

Family

ID=43627165

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/075682 WO2011022905A1 (zh) 2009-08-25 2009-12-17 单环网络拓扑重建方法及系统

Country Status (4)

Country Link
US (1) US20120155245A1 (zh)
EP (1) EP2472795B1 (zh)
CN (1) CN101997735A (zh)
WO (1) WO2011022905A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102916884A (zh) * 2012-10-26 2013-02-06 江苏华丽网络工程有限公司 一种交换设备的拓扑发现方法
CN113946141A (zh) * 2020-07-16 2022-01-18 青岛海尔工业智能研究院有限公司 一种生产示范线的网络系统及控制方法

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102761451A (zh) * 2012-08-09 2012-10-31 武汉迈威实达软件有限公司 一种基于rstp改进型单环路冗余备份的实现
CN104283753A (zh) * 2013-07-10 2015-01-14 恒启电子(苏州)有限公司 一种支持无线等多种点到点介质的环网技术
CN103840965B (zh) * 2013-10-18 2017-02-08 深圳市新格林耐特通信技术有限公司 一种增强rstp中故障快速收敛的方法
GB2524749B (en) * 2014-03-31 2018-12-19 Metaswitch Networks Ltd Spanning tree protocol
CN105515809A (zh) * 2014-09-28 2016-04-20 中兴通讯股份有限公司 一种软件定义网络实现方法和主控制器
JP6554405B2 (ja) * 2015-11-26 2019-07-31 株式会社日立製作所 リングネットワークシステム、及び、ネットワークノード
CN107995111B (zh) * 2016-10-26 2021-08-06 中兴通讯股份有限公司 业务转发、链路变化通告方法、汇聚设备及接入设备
EP3568957B1 (en) * 2017-01-11 2021-03-03 Unify Patente GmbH & Co. KG Method of operating a unit in a daisy chain, communication unit and a system including a plurality of communication units
CN108933720B (zh) * 2017-05-25 2021-11-02 中兴通讯股份有限公司 环形系统的信息处理方法、装置、系统及存储介质
US11025537B2 (en) * 2017-12-04 2021-06-01 Is5 Communications, Inc. Multiple RSTP domain separation
JP7172721B2 (ja) 2019-02-26 2022-11-16 株式会社明電舎 フレーム転送の制御方式
CN111585797B (zh) * 2020-04-27 2023-04-07 深圳震有科技股份有限公司 以太网链路切换方法、装置、设备及计算机可读存储介质
TWI733560B (zh) * 2020-08-13 2021-07-11 瑞昱半導體股份有限公司 網路交換器及其網路交換器系統
CN114172855A (zh) * 2020-08-19 2022-03-11 瑞昱半导体股份有限公司 网络交换器及其网络交换器系统
JP2022034686A (ja) * 2020-08-19 2022-03-04 アズビル株式会社 ネットワーク装置およびネットワーク構成判定方法
CN114978939B (zh) * 2022-06-10 2023-05-30 中煤科工重庆设计研究院(集团)有限公司 一种网络链路质量的检测方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1812361A (zh) * 2006-01-23 2006-08-02 杭州华为三康技术有限公司 快速环网保护方法及系统
CN1870558A (zh) * 2006-05-17 2006-11-29 杭州华为三康技术有限公司 弹性分组环中相交环的环路检测方法和系统
US20090207726A1 (en) * 2008-02-14 2009-08-20 Graeme Thomson System and method for network recovery from multiple link failures

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3715501B2 (ja) * 2000-03-10 2005-11-09 アンリツ株式会社 スパニングツリー用ブリッジ及びそれを用いた経路変更方法
US6717922B2 (en) * 2002-03-04 2004-04-06 Foundry Networks, Inc. Network configuration protocol and method for rapid traffic recovery and loop avoidance in ring topologies
JP3635268B2 (ja) * 2002-05-23 2005-04-06 アンリツ株式会社 ブリッジ及びそれを用いた経路変更方法
JP3729265B2 (ja) * 2002-08-22 2005-12-21 日本電気株式会社 ネットワークシステム、スパニングツリー構成方法、スパニングツリー構成ノード、及びスパニングツリー構成プログラム
JP4418252B2 (ja) * 2004-02-17 2010-02-17 富士通株式会社 リング型光ネットワークにおける信号送受信方法及び同ネットワークに用いられる光伝送ノード
JP2005269059A (ja) * 2004-03-17 2005-09-29 Fujitsu Ltd データ中継装置、データ中継方法およびデータ中継プログラム
JP2005318086A (ja) * 2004-04-27 2005-11-10 Fujitsu Ltd 伝送装置
CN101189838B (zh) * 2005-03-31 2014-09-10 日本电气株式会社 环形网络系统、故障恢复方法和节点
US7653011B2 (en) * 2005-05-31 2010-01-26 Cisco Technology, Inc. Spanning tree protocol for wireless networks
US7778205B2 (en) * 2005-08-30 2010-08-17 Cisco Technology, Inc. System and method for implementing virtual ports within ring networks
EP1935145B1 (en) * 2005-10-11 2011-12-21 Telefonaktiebolaget LM Ericsson (publ) Method of generating spanning trees
US7760668B1 (en) * 2006-06-20 2010-07-20 Force 10 Networks, Inc. Self-reconfiguring spanning tree
US7903586B2 (en) * 2006-11-01 2011-03-08 Alcatel Lucent Ring rapid multiple spanning tree protocol system and method
TWI379547B (en) * 2008-10-15 2012-12-11 Redundancy method of ring network

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1812361A (zh) * 2006-01-23 2006-08-02 杭州华为三康技术有限公司 快速环网保护方法及系统
CN1870558A (zh) * 2006-05-17 2006-11-29 杭州华为三康技术有限公司 弹性分组环中相交环的环路检测方法和系统
US20090207726A1 (en) * 2008-02-14 2009-08-20 Graeme Thomson System and method for network recovery from multiple link failures

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2472795A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102916884A (zh) * 2012-10-26 2013-02-06 江苏华丽网络工程有限公司 一种交换设备的拓扑发现方法
CN102916884B (zh) * 2012-10-26 2015-12-09 唐稳杰 一种交换设备的拓扑发现方法
CN113946141A (zh) * 2020-07-16 2022-01-18 青岛海尔工业智能研究院有限公司 一种生产示范线的网络系统及控制方法
CN113946141B (zh) * 2020-07-16 2022-09-06 卡奥斯工业智能研究院(青岛)有限公司 一种生产示范线的网络系统及控制方法

Also Published As

Publication number Publication date
EP2472795A1 (en) 2012-07-04
CN101997735A (zh) 2011-03-30
US20120155245A1 (en) 2012-06-21
EP2472795A4 (en) 2014-07-23
EP2472795B1 (en) 2018-03-14

Similar Documents

Publication Publication Date Title
WO2011022905A1 (zh) 单环网络拓扑重建方法及系统
US7233991B2 (en) Self-healing tree network
US8520508B2 (en) Spanning tree ring protocol
US7924702B2 (en) Method for reconfiguring a communication network
JP3715501B2 (ja) スパニングツリー用ブリッジ及びそれを用いた経路変更方法
JP5566463B2 (ja) コンピュータネットワークにおけるデータ転送の技術
US6934263B1 (en) Spanning tree with protocol for bypassing port state transition timers
JP3635268B2 (ja) ブリッジ及びそれを用いた経路変更方法
WO2014026512A1 (zh) 一种快速生成树协议应用于单环网拓扑的处理方法
CN101610221B (zh) 一种stp切换时ip单播平滑切换的方法及装置
US20060253561A1 (en) Method and apparatus for fast re-configuration of a network topology
US9906441B2 (en) Rapid flood processing
WO2012048585A1 (zh) 切换方法和路由器
WO2012162946A1 (zh) 一种报文处理方法及系统
WO2003084129A1 (fr) Procede de collecte de donnees de topologie de reseau
US8902793B2 (en) Method and system for a low-complexity spanning tree algorithm in communication networks
US20130100854A1 (en) Vpls over multi-chassis trunk
US6724734B1 (en) Creating a spanning tree of a network including clusters
US20050237943A1 (en) Transmission device
CN108833247B (zh) 链路故障恢复后的再切换方法及装置
CN102075374A (zh) 单环网络拓扑重建方法及系统
JP6113157B2 (ja) 通信ネットワークにおいてルーターをスイッチオフする方法と、この方法を実施するルーター
KR100917299B1 (ko) 스패닝 트리에서 링크집합과의 연동을 지원하는 루프 제거방법
JP2007142751A (ja) ブリッジ及びそれを用いた経路変更方法
JP5427665B2 (ja) スパニングツリーの再構成方法および通信装置

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: 09848637

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13392199

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2009848637

Country of ref document: EP