[go: up one dir, main page]

CN100417139C - A method for uninterrupted forwarding of multicast data - Google Patents

A method for uninterrupted forwarding of multicast data Download PDF

Info

Publication number
CN100417139C
CN100417139C CNB2005101020840A CN200510102084A CN100417139C CN 100417139 C CN100417139 C CN 100417139C CN B2005101020840 A CNB2005101020840 A CN B2005101020840A CN 200510102084 A CN200510102084 A CN 200510102084A CN 100417139 C CN100417139 C CN 100417139C
Authority
CN
China
Prior art keywords
port
multicast
root
forwarding
neighbor discovery
Prior art date
Legal status (The legal status 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 status listed.)
Expired - Fee Related
Application number
CNB2005101020840A
Other languages
Chinese (zh)
Other versions
CN1859288A (en
Inventor
牛乐宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CNB2005101020840A priority Critical patent/CN100417139C/en
Publication of CN1859288A publication Critical patent/CN1859288A/en
Application granted granted Critical
Publication of CN100417139C publication Critical patent/CN100417139C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Small-Scale Networks (AREA)

Abstract

本发明公开了一种组播数据不间断转发的方法,涉及计算机网络通信技术领域。当非根桥三层设备上生成树协议的根端口切换时,非根桥三层设备将新的根端口设置为组播转发端口,并通过新的根端口向根桥三层设备的指定端口发送邻居发现协议报文;根桥三层设备收到邻居发现协议报文时,将指定端口设置为组播转发端口,通过该指定端口向新的根端口回应邻居发现协议报文;非根桥三层设备收到根桥三层设备回应的邻居发现协议报文后,新的根端口与所述指定端口之间的链路建立,组播数据通过该链路继续转发。本发明的方法能够缩短因根端口切换时产生的组播转发阻断的时间间隔,使时间间隔降低到10毫秒级,满足组播业务实时性的要求,提高组播转发的效率。

Figure 200510102084

The invention discloses a method for uninterrupted forwarding of multicast data, and relates to the technical field of computer network communication. When the root port of the spanning tree protocol on the non-root bridge layer 3 device is switched, the non-root bridge layer 3 device sets the new root port as the multicast forwarding port, and transmits to the designated port of the root bridge layer 3 device through the new root port. Send Neighbor Discovery Protocol packets; when the root bridge Layer 3 device receives Neighbor Discovery Protocol packets, it sets the designated port as a multicast forwarding port, and responds to the new root port through the designated port. Neighbor Discovery Protocol packets; non-root bridge After the layer-3 device receives the neighbor discovery protocol message responded by the layer-3 device of the root bridge, the link between the new root port and the specified port is established, and the multicast data is forwarded continuously through the link. The method of the invention can shorten the time interval of multicast forwarding blocking caused by root port switching, reduce the time interval to 10 milliseconds, meet the real-time requirement of multicast business, and improve the efficiency of multicast forwarding.

Figure 200510102084

Description

一种组播数据不间断转发的方法 A method for uninterrupted forwarding of multicast data

[技术领域][technical field]

本发明涉及计算机网络通信技术领域,尤其涉及一种保持组播数据不间断转发的方法。The invention relates to the technical field of computer network communication, in particular to a method for maintaining uninterrupted forwarding of multicast data.

[背景技术][Background technique]

如图1所示,是计算机通信网络的基本结构图,用户的流量通过接入网接入三层交换机,多个三层交换机将流量向上一级三层交换机汇聚(如图中的虚线框所示),通过三层交换机的汇聚作用,最后将用户的流量通过核心网送达Internet业务提供商,Internet业务提供商为用户提供的业务按照相反的方向传递到用户,从而实现业务的端到端连接。其中,这些三层交换机的连接形成了接入层的汇聚网络,一般来说,三层交换机连接的VLAN(虚拟局域网)划分为两种形式:As shown in Figure 1, it is a basic structural diagram of a computer communication network. User traffic is connected to a Layer 3 switch through the access network, and multiple Layer 3 switches converge the traffic to the upper Layer 3 switch (as indicated by the dotted box in the figure). As shown), through the aggregation function of the three-layer switch, the user's traffic is finally delivered to the Internet service provider through the core network, and the service provided by the Internet service provider for the user is transmitted to the user in the opposite direction, thereby realizing the end-to-end service connect. Among them, the connection of these three-layer switches forms the aggregation network of the access layer. Generally speaking, the VLAN (virtual local area network) connected by the three-layer switch is divided into two forms:

1、通过二层汇聚,所有三层交换机的端口都配置成相同的VLAN,所有的流量通过二层以太网交换的方式汇聚到核心网。1. Through Layer 2 convergence, all ports of Layer 3 switches are configured as the same VLAN, and all traffic is converged to the core network through Layer 2 Ethernet switching.

2、通过三层汇聚,三层交换机汇聚流量的入VLAN与出VLAN配置不同,也就是说,三层交换机的入口和出口配置不同的VLAN,这些VLAN间要达到互通,只能通过三层的IP互通,流量通过三层IP交换的方式汇聚到核心网。2. Through the three-layer aggregation, the configuration of the inbound VLAN and the outbound VLAN of the traffic aggregated by the three-layer switch is different. IP intercommunication, the traffic is converged to the core network through three-layer IP switching.

为了保证链路的可靠性,对接的某一个三层交换机可能将多个物理端口(例如N个,N≥2)配置在同一个VLAN内(多端口VLAN),同时另一个交换机也有多个端口(N个)配置在这个VLAN内,这两个交换机的N个物理端口一一对接。因为组播数据在二层以太网链路的发送方式是广播发送的,因此支持组播业务的应用,在多端口VLAN对接的情况下,组播数据会在对接的以太网链路上产生广播风暴,为了防止二层以太网链路上的广播风暴,应该在三层交换机的所有端口上运行生成树协议,如RSTP(快速生成树协议)、MSTP(多生成树协议)。如果三层交换机是通过三层汇聚方式对接的,为支持组播转发,三层交换机上应该启动组播路由协议,如PIM(协议无关多播)协议,交换机之间通过交换组播路由协议信息在整个网络上生成组播转发树,在每一个交换机内部生成组播转发表,指导组播数据的转发,同时交换机之间通过周期性的发送握手报文(如PIM的HELLO报文)来保持连通性。在多端口VLAN的情况下,VLAN下面的端口因为运行了生成树协议,在一个VLAN接口包含的所有物理端口上,同一时刻只能从唯一的端口上收到组播协议报文,并且组播流量只在收到组播路由协议的端口上发送,从而达到按照端口精确转发组播流量的目的。In order to ensure the reliability of the link, a Layer 3 switch that is interconnected may configure multiple physical ports (for example, N, N≥2) in the same VLAN (multi-port VLAN), and another switch also has multiple ports. (N) are configured in this VLAN, and the N physical ports of the two switches are connected one by one. Because multicast data is sent by broadcast on the Layer 2 Ethernet link, it supports the application of multicast services. In the case of multi-port VLAN interconnection, multicast data will be broadcast on the interconnected Ethernet link. Storm, in order to prevent broadcast storms on Layer 2 Ethernet links, spanning tree protocols, such as RSTP (Rapid Spanning Tree Protocol) and MSTP (Multiple Spanning Tree Protocol), should be run on all ports of Layer 3 switches. If the Layer 3 switches are connected through Layer 3 aggregation, in order to support multicast forwarding, a multicast routing protocol, such as PIM (Protocol Independent Multicast) protocol, should be enabled on the Layer 3 switch, and the switches exchange multicast routing protocol information A multicast forwarding tree is generated on the entire network, and a multicast forwarding table is generated inside each switch to guide the forwarding of multicast data. At the same time, switches periodically send handshake messages (such as PIM HELLO messages) to maintain connectivity. In the case of multi-port VLAN, because the ports under the VLAN run the spanning tree protocol, on all the physical ports included in a VLAN interface, they can only receive multicast protocol packets from the only port at the same time, and the multicast Traffic is only sent on the port that receives the multicast routing protocol, so as to achieve the purpose of accurately forwarding multicast traffic according to the port.

三层交换机在VLAN下面的所有端口上实时侦听是否收到组播协议报文,如果端口侦听到组播协议报文,就将这个端口设置为组播转发端口,而将没有侦听组播协议报文的端口设置为阻塞组播转发端口;如果组播转发端口在一段时间内没有收到组播协议报文,组播转发端口就会变成阻塞组播转发端口,由此保证了组播数据在端口上能够被精确转发。The Layer 3 switch listens in real time on all ports under the VLAN to see if the multicast protocol message is received. If the port detects the multicast protocol message, it will set this port as a multicast forwarding port, and there will be no listening group. The port of the multicast protocol message is set as a blocking multicast forwarding port; if the multicast forwarding port does not receive a multicast protocol message within a period of time, the multicast forwarding port will become a blocked multicast forwarding port, thereby ensuring Multicast data can be accurately forwarded on the port.

如图2所示,为三层交换机对接的示意图。三层交换机L3(A)、三层交换机L3(B)上运行组播路由协议,两者通过相同VLAN号的三层接口对接,三层交换机L3(A)的P1、P2端口、三层交换机L3(B)的P3、P4端口是属于对接VLAN的以太网端口,这些端口都配置在同一个VLAN内,P1端口与P3端口物理上相接,P2端口与P4端口物理上相接。As shown in FIG. 2 , it is a schematic diagram of layer 3 switch interconnection. Layer 3 switch L3(A) and Layer 3 switch L3(B) run the multicast routing protocol, and they are interconnected through Layer 3 interfaces with the same VLAN number. The P1 and P2 ports of Layer 3 switch L3(A) The P3 and P4 ports of L3(B) are Ethernet ports connected to the VLAN. These ports are configured in the same VLAN. The P1 port is physically connected to the P3 port, and the P2 port is physically connected to the P4 port.

在P1、P2、P3、P4端口上运行生成树协议,设置交换机L3(A)的桥ID小于交换机L3(B)的桥ID,P1对应的端口ID小于P2对应的端口ID,P3对应的端口ID小于P4对应的端口ID,通过生成树协议机制,选举交换机L3(A)为根桥,P1、P2是生成树协议的指定端口,端口的生成树协议状态为Forwarding(转发),可以接收转发数据;交换机L3(B)为非根桥,P3为根端口,端口的生成树协议状态为Forwarding(转发),可以接收转发数据,P4为可选端口,端口的生成树协议状态为Discarding(丢弃),不能接收转发数据。Run the spanning tree protocol on ports P1, P2, P3, and P4, set the bridge ID of switch L3 (A) to be smaller than the bridge ID of switch L3 (B), the port ID corresponding to P1 is smaller than the port ID corresponding to P2, and the port corresponding to P3 ID is smaller than the port ID corresponding to P4. Through the spanning tree protocol mechanism, the switch L3 (A) is elected as the root bridge. P1 and P2 are designated ports of the spanning tree protocol. The spanning tree protocol state of the port is Forwarding (forwarding), which can receive and forward data; switch L3 (B) is a non-root bridge, P3 is a root port, the spanning tree protocol state of the port is Forwarding (forwarding), can receive and forward data, P4 is an optional port, and the spanning tree protocol state of the port is Discarding (discarding) ), cannot receive forwarded data.

交换机L3(A)、L3(B)之间相互发送组播协议报文,交换机L3(A)只能在P1端口上接收到交换机L3(B)从P3端口上发送的组播协议报文,交换机L3(B)只能在P3端口上接收到交换机L3(A)从P1端口上发送的组播协议报文。因此交换机L3(A)、L3(B)分别将P1、P3端口设置为组播转发端口,将P2、P4端口设置为阻塞组播转发端口。The switches L3(A) and L3(B) send multicast protocol packets to each other, and the switch L3(A) can only receive the multicast protocol packets sent by the switch L3(B) from the P3 port on the P1 port. The switch L3(B) can only receive the multicast protocol packet sent by the switch L3(A) from the P1 port on the P3 port. Therefore, switches L3 (A) and L3 (B) respectively set P1 and P3 ports as multicast forwarding ports, and set P2 and P4 ports as blocking multicast forwarding ports.

在生成树协议的运行过程中,如果交换机L3(B)的根端口从P3改变为P4,这种根端口切换操作是10毫秒级的,切换的结果是P3端口的生成树协议状态由Forwarding转变为Discarding,P4端口的生成树协议状态由Discarding转变为Forwarding,但是,在这种情况P4端口无法立刻转发组播数据,从而导致交换机L3(A)、L3(B)间组播数据转发被阻断。因为P4端口还没有收到交换机间互相发送的定期握手报文,所以P4端口还是阻塞组播转发端口;一般情况下,交换机间互相发送的定期握手报文的发送时间间隔是分钟级的,在根端口切换时,交换机L3(B)并不立刻从P4端口发送握手报文,而是在交换机L3(B)规定的下一个握手报文的发送周期到来时,才会向交换机L3(A)发送握手报文,当交换机L3(A)上的P2端口第一次接收到交换机L3(B)从P4端口发送的握手报文时,交换机L3(A)将P2端口从阻塞组播转发端口改变为组播转发端口,然后,在交换机L3(A)规定的下一个握手报文的发送周期到来时,才会向交换机L3(B)发送握手报文,交换机L3(B)在P4端口第一次收到握手报文时,交换机L3(B)将P4端口从阻塞组播转发端口改变为组播转发端口,到此为止,交换机L3(A)、L3(B)间才恢复了组播数据转发。由于P1端口在一段时间内没有收到周期性握手报文,P1端口就会从组播转发端口改变为阻塞组播转发端口。这里需要说明的是,在根端口切换时,并不一定是先由P4端口向P2端口发送握手报文,也有可能是P2端口先向P4端口发送握手报文,换言之,P2、P4端口的先后发送握手报文是随机的。但无论是P2端口还是P4端口先发送握手报文,交换机L3(A)、L3(B)之间对组播数据转发都会达到分钟级的阻断,这不依赖于系统处理器的速度,而是依赖于协议规定的时间。During the operation of the spanning tree protocol, if the root port of the switch L3 (B) changes from P3 to P4, this root port switching operation is at the level of 10 milliseconds, and the result of the switching is that the spanning tree protocol state of the P3 port changes from Forwarding For Discarding, the spanning tree protocol status of the P4 port changes from Discarding to Forwarding. However, in this case, the P4 port cannot forward the multicast data immediately, resulting in the blocking of multicast data forwarding between switches L3(A) and L3(B). broken. Because the P4 port has not received the regular handshake message sent between the switches, the P4 port is still blocking the multicast forwarding port; under normal circumstances, the sending interval of the periodic handshake message sent between the switches is at the minute level. When the root port is switched, the switch L3(B) does not immediately send the handshake message from the P4 port, but will send the handshake message to the switch L3(A) when the next sending cycle of the handshake message stipulated by the switch L3(B) arrives. Send a handshake message, when the P2 port on the switch L3(A) receives the handshake message sent by the switch L3(B) from the P4 port for the first time, the switch L3(A) changes the P2 port from the blocking multicast forwarding port The port is forwarded for multicast, and then, when the sending cycle of the next handshake message stipulated by the switch L3(A) arrives, the handshake message will be sent to the switch L3(B), and the switch L3(B) is the first on the P4 port When the handshake message is received for the first time, the switch L3(B) changes the P4 port from the blocking multicast forwarding port to the multicast forwarding port, so far, the multicast data between the switches L3(A) and L3(B) is restored Forward. Since the P1 port does not receive a periodic handshake packet within a period of time, the P1 port will change from a multicast forwarding port to a blocking multicast forwarding port. What needs to be explained here is that when the root port is switched, it is not necessarily that the P4 port sends the handshake message to the P2 port first, and it may also be that the P2 port sends the handshake message to the P4 port first. In other words, the order of the P2 and P4 ports Sending handshake messages is random. However, regardless of whether the P2 port or the P4 port sends the handshake message first, the forwarding of multicast data between the switches L3(A) and L3(B) will be blocked at the minute level, which does not depend on the speed of the system processor. is dependent on the time specified in the agreement.

因此,目前的技术方案在生成树协议根端口切换时,交换机之间就会产生分钟级的组播数据阻断,对于在三层交换机上开展实时组播业务来说,这种分钟级的组播数据阻断是不能容许的。Therefore, in the current technical solution, when the root port of the spanning tree protocol is switched, minute-level multicast data blocking will be generated between the switches. Broadcast data blocking is not allowed.

[发明内容][Content of the invention]

本发明要解决的技术问题是提供一种组播数据不间断转发的方法,在三层交换机之间通过包含多物理端口的VLAN相对接的情况下,且在VLAN接口上运行组播路由协议,在VLAN的所有端口上运行生成树协议,当生成树协议的根端口变化时,能够保持组播数据不间断转发,缩短因根端口切换时产生的组播转发阻断的时间间隔,提高组播转发的效率。The technical problem to be solved in the present invention is to provide a method for uninterrupted forwarding of multicast data, in the case of interconnection between Layer 3 switches through VLANs containing multiple physical ports, and run a multicast routing protocol on the VLAN interface, Run the spanning tree protocol on all ports of the VLAN. When the root port of the spanning tree protocol changes, it can keep multicast data forwarding uninterruptedly, shorten the time interval of multicast forwarding blocking caused by root port switching, and improve multicast Forwarding efficiency.

本发明是通过以下的技术方案来实现的:The present invention is achieved through the following technical solutions:

一种组播数据不间断转发的方法,当非根桥三层设备上生成树协议的根端口切换时,包括以下步骤:A method for uninterrupted forwarding of multicast data, when the root port of the spanning tree protocol on the non-root bridge layer 3 device is switched, the steps include the following steps:

101、非根桥三层设备将新的根端口设置为组播转发端口,所述非根桥三层设备不等待邻居发现协议报文的发送周期到来,通过新的根端口向根桥三层设备的指定端口发送邻居发现协议报文;101. The layer-3 device of the non-root bridge sets the new root port as a multicast forwarding port, and the layer-3 device of the non-root bridge does not wait for the sending period of the neighbor discovery protocol message to arrive, and transmits the message to the layer-3 root port through the new root port. The designated port of the device sends neighbor discovery protocol packets;

102、根桥三层设备收到所述邻居发现协议报文时,将所述指定端口设置为组播转发端口,所述根桥三层设备不等待邻居发现协议报文的发送周期到来,通过该指定端口向所述新的根端口回应邻居发现协议报文;102. When the third-layer device of the root bridge receives the neighbor discovery protocol message, it sets the designated port as a multicast forwarding port, and the third-layer device of the root bridge does not wait for the sending period of the neighbor discovery protocol message to arrive, and passes The designated port responds to the neighbor discovery protocol message to the new root port;

103、非根桥三层设备收到根桥三层设备回应的邻居发现协议报文后,所述新的根端口与所述指定端口之间的链路建立,组播数据通过该链路继续转发。103. After the third-layer device of the non-root bridge receives the neighbor discovery protocol message responded by the third-layer device of the root bridge, the link between the new root port and the designated port is established, and the multicast data continues through this link. Forward.

在步骤101之前还包括:当非根桥三层设备上生成树协议的根端口切换时,新的根端口和旧的根端口的生成树协议状态分别为转发和丢弃。Before step 101, it also includes: when the root port of the spanning tree protocol on the non-root bridge three-layer device is switched, the spanning tree protocol states of the new root port and the old root port are forwarding and discarding respectively.

步骤101中,当非根桥三层设备将新的根端口设置为组播转发端口时,将旧的根端口设置为阻塞组播转发端口。In step 101, when the non-root bridge layer-3 device sets a new root port as a multicast forwarding port, it sets an old root port as a blocking multicast forwarding port.

步骤101中,非根桥三层设备首先超时其内部的周期性邻居发现协议报文发送定时器,在所述周期性邻居发现协议报文发送定时器超时后,通过新的根端口向根桥三层设备的指定端口发送邻居发现协议报文。In step 101, the non-root bridge three-layer device first overtimes its internal periodic Neighbor Discovery Protocol packet sending timer, and after the periodic Neighbor Discovery Protocol packet sending timer expires, it sends a new root port to the root bridge. The designated port of the layer-3 device sends neighbor discovery protocol packets.

步骤102中,根桥三层设备首先超时其内部的周期性邻居发现协议报文发送定时器,在所述周期性邻居发现协议报文发送定时器超时后,通过所述指定端口向所述新的根端口回应邻居发现协议报文。In step 102, the three-layer device of the root bridge first times out its internal periodic Neighbor Discovery Protocol packet sending timer, and after the periodic Neighbor Discovery Protocol packet sending timer expires, it sends a message to the new The root port of the node responds to neighbor discovery protocol packets.

本发明的改进在于:所述三层设备为三层交换机或者路由器。The improvement of the present invention is that: the three-layer equipment is a three-layer switch or router.

本发明的改进在于:所述邻居发现协议报文是协议无关多播的你好报文或者参加报文。The improvement of the present invention is that: the neighbor discovery protocol message is a protocol-independent multicast hello message or a join message.

本发明的改进在于:所述生成树协议为快速生成树协议或者多生成树协议。The improvement of the present invention is that: the spanning tree protocol is rapid spanning tree protocol or multiple spanning tree protocol.

本发明中,当生成树协议的根端口变化时,采用上述的技术方案能够保持组播数据不间断转发,缩短因根端口切换时产生的组播转发阻断的时间间隔,使时间间隔从分钟级降低到10毫秒级,满足组播业务实时性的要求,提高组播转发的效率。In the present invention, when the root port of the spanning tree protocol changes, the above-mentioned technical scheme can keep the multicast data forwarding uninterruptedly, shorten the time interval of multicast forwarding blocking caused by root port switching, and make the time interval from minute to minute The level is reduced to 10 milliseconds, which meets the real-time requirements of multicast services and improves the efficiency of multicast forwarding.

[附图说明][Description of drawings]

图1是计算机通信网络的基本结构图。Figure 1 is a basic structural diagram of a computer communication network.

图2是两个三层交换机对接示意图。Fig. 2 is a schematic diagram of interconnection between two layer 3 switches.

图3是本发明中实现组播数据不间断转发的方法流程图。Fig. 3 is a flowchart of a method for realizing uninterrupted forwarding of multicast data in the present invention.

[具体实施方式][Detailed ways]

下面结合附图和实施例对本发明进行进一步阐述:The present invention is further elaborated below in conjunction with accompanying drawing and embodiment:

本发明中三层交换机之间通过包含多物理端口的VLAN相对接,在对接的VLAN接口上运行组播路由协议,在对接的该VLAN包含的所有以太网端口上运行生成树协议;端口的状态定义为两种:组播转发状态、阻塞组播转发状态。In the present invention, the three-layer switch is docked by VLANs that include multiple physical ports, and the multicast routing protocol is run on the docked VLAN interface, and the spanning tree protocol is run on all Ethernet ports included in the docked VLAN; the state of the port It is defined as two types: multicast forwarding state and blocked multicast forwarding state.

如图2所示,两个三层交换机对接,三层交换机L3(A)、三层交换机L3(B)上运行组播路由协议,两者通过相同VLAN号的三层接口对接,三层交换机L3(A)的P1、P2端口、三层交换机L3(B)的P3、P4端口是属于对接VLAN的以太网端口,这些端口都配置在同一个VLAN内,P1端口与P3端口物理上相接,P2端口与P4端口物理上相接。As shown in Figure 2, two Layer 3 switches are interconnected. The Layer 3 switch L3(A) and the Layer 3 switch L3(B) run the multicast routing protocol. The two are interconnected through the Layer 3 interface with the same VLAN number. The Layer 3 switch The P1 and P2 ports of L3(A) and the P3 and P4 ports of L3(B) switch are Ethernet ports connected to the VLAN. These ports are all configured in the same VLAN, and the P1 port and the P3 port are physically connected. , the P2 port is physically connected to the P4 port.

在P1、P2、P3、P4端口上运行生成树协议,设置三层交换机L3(A)的桥ID小于三层交换机L3(B)的桥ID,P1对应的端口ID小于P2对应的端口ID,P3对应的端口ID小于P4对应的端口ID,通过生成树协议机制,选举三层交换机L3(A)为根桥,P1、P2是生成树协议的指定端口,端口的生成树协议状态为Forwarding(转发),可以接收转发数据;三层交换机L3(B)为非根桥,P3为根端口,端口的生成树协议状态为Forwarding(转发),可以接收转发数据,P4为可选端口,端口的生成树协议状态为Discarding(丢弃),不能接收转发数据。Run the spanning tree protocol on the P1, P2, P3, and P4 ports, set the bridge ID of the three-layer switch L3 (A) to be smaller than the bridge ID of the three-layer switch L3 (B), and the port ID corresponding to P1 is smaller than the port ID corresponding to P2, The port ID corresponding to P3 is smaller than the port ID corresponding to P4. Through the spanning tree protocol mechanism, the three-layer switch L3 (A) is elected as the root bridge. P1 and P2 are designated ports of the spanning tree protocol. The spanning tree protocol state of the ports is Forwarding ( Forwarding), can receive forwarding data; Layer 3 switch L3(B) is a non-root bridge, P3 is a root port, the spanning tree protocol state of the port is Forwarding (forwarding), can receive forwarding data, P4 is an optional port, the port’s The spanning tree protocol status is Discarding (discarding), and forwarding data cannot be received.

在正常工作的情况下,三层交换机L3(A)、三层交换机L3(B)上都分别启动了周期性握手报文发送定时器,假设周期为m秒(m>0秒),定时器周期性的经过m秒超时,三层交换机L3(A)和三层交换机L3(B)相互发送组播握手报文,只有这样才能保证两个三层交换机之间的可靠互通。由于组播握手报文是在P1端口和P3端口之间相互发送的,因此P1、P3端口设置为组播转发状态,而P2、P4端口设置为阻塞组播转发状态。Under normal working conditions, the three-layer switch L3(A) and the three-layer switch L3(B) respectively start the periodic handshake message sending timer, assuming that the period is m seconds (m>0 seconds), the timer Periodically through the timeout of m seconds, the three-layer switch L3 (A) and the three-layer switch L3 (B) send multicast handshake messages to each other, only in this way can the reliable intercommunication between the two three-layer switches be guaranteed. Since the multicast handshake message is sent between the P1 port and the P3 port, the P1 and P3 ports are set to the multicast forwarding state, and the P2 and P4 ports are set to the blocking multicast forwarding state.

这里需要补充的是组播握手报文可以是PIM的HELLO(你好)报文,,也可以是JOIN(参加)报文,但不管是哪一种报文,都可以统称为邻居发现协议报文,交换机之间互发邻居发现协议报文,其目的是确认对方的可达性,使得组播流量能够顺利到达对方。What needs to be added here is that the multicast handshake message can be a PIM HELLO (hello) message, or a JOIN (join) message, but no matter what kind of message it is, it can be collectively referred to as a neighbor discovery protocol message. In this document, switches send Neighbor Discovery Protocol messages to each other to confirm the reachability of each other, so that multicast traffic can reach each other smoothly.

如图3所示,为实现组播数据不间断转发的方法流程图,其具体过程为:As shown in Figure 3, in order to realize the flow chart of the method for uninterrupted forwarding of multicast data, its specific process is:

第一步:当P1端口与P3端口之间发生链路故障时,组播流转发中断,三层交换机L3(B)检测到这一故障信息时,三层交换机L3(B)上就会发生根端口切换,即:根端口由P3端口(旧的根端口)变成P4端口(新的根端口);三层交换机L3(B)立刻将P4端口从阻塞组播转发状态改变为组播转发状态,同时将P3端口从组播转发状态改变为阻塞组播转发状态。此时,P3端口的生成树协议状态为Discarding(丢弃),不能接收转发数据,P4端口的生成树协议状态为Forwarding(转发),可以接收转发数据。Step 1: When a link failure occurs between the P1 port and the P3 port, the forwarding of the multicast stream is interrupted. When the layer-3 switch L3(B) detects this failure information, a failure occurs on the layer-3 switch L3(B). Root port switching, that is: the root port changes from P3 port (old root port) to P4 port (new root port); Layer 3 switch L3(B) immediately changes P4 port from blocking multicast forwarding state to multicast forwarding At the same time, change the P3 port from the multicast forwarding state to the blocked multicast forwarding state. At this time, the spanning tree protocol state of the P3 port is Discarding (discarding), unable to receive forwarded data, and the spanning tree protocol state of the P4 port is Forwarding (forwarding), capable of receiving forwarded data.

第二步:三层交换机L3(B)立刻超时周期性握手报文发送定时器,不等待组播握手报文的发送周期到来,在P4端口上立刻向P2端口发送组播握手报文。Second step: the three-layer switch L3 (B) immediately overtimes the periodic handshake message sending timer, does not wait for the arrival of the sending period of the multicast handshake message, and sends the multicast handshake message to the P2 port immediately on the P4 port.

第三步:当三层交换机L3(A)的P2端口(P2端口的生成树协议状态为Forwarding(转发),可以接收转发数据)上第一次收到组播握手报文时,P2端口立刻从阻塞组播转发状态改变为组播转发状态;三层交换机L3(A)立刻超时周期性握手报文发送定时器,不等待组播握手报文的发送周期到来,在P2端口上立刻向P4端口发送组播握手报文。Step 3: When the P2 port of the three-layer switch L3 (A) (the spanning tree protocol status of the P2 port is Forwarding (forwarding), which can receive and forward data) receives the multicast handshake message for the first time, the P2 port immediately Change from the blocked multicast forwarding state to the multicast forwarding state; the three-layer switch L3 (A) immediately times out the periodic handshake message sending timer, does not wait for the arrival of the multicast handshake message sending period, and sends P4 immediately on the P2 port The port sends a multicast handshake message.

第四步:P4端口接收到组播握手报文,表明三层交换机L3(B)与三层交换机L3(A)都能够相互到达,此时,P2、P4端口都设置为组播转发状态,P2、P4之间链路建立,组播流量通过这条链路实现不间断转发。Step 4: The P4 port receives the multicast handshake message, indicating that the three-layer switch L3 (B) and the three-layer switch L3 (A) can reach each other. At this time, the P2 and P4 ports are all set to the multicast forwarding state. The link between P2 and P4 is established, and multicast traffic is forwarded continuously through this link.

本发明实现不间断转发组播流量的方法,避开了三层交换机之间相互发送组播握手报文所存在的周期性时间间隔,当非根桥三层交换机检测到链路层故障而发生以太网端口切换(根端口发生切换)时,非根桥三层交换机立刻超时组播握手报文周期定时器,通过新的根端口主动发送组播握手报文,而根桥交换机的任何一个以太网端口第一次收到组播握手报文时,根桥三层交换机也立刻超时组播握手报文周期定时器,向非根桥交换机发送组播握手报文,这样就加快了根桥交换机与非根桥交换机之间的链路建立,减少组播业务流中断的时间,提高了转发效率。The method of the present invention realizes the uninterrupted forwarding of multicast traffic, avoids the periodic time interval between the three-layer switches that send multicast handshake messages to each other, and occurs when the non-root bridge three-layer switch detects a link layer failure. When the Ethernet port is switched (the root port is switched), the non-root bridge Layer 3 switch immediately times out the multicast handshake message cycle timer, and actively sends the multicast handshake message through the new root port, while any Ethernet port of the root bridge switch When the network port receives the multicast handshake message for the first time, the root bridge Layer 3 switch also immediately times out the multicast handshake message period timer, and sends the multicast handshake message to the non-root bridge switch, which speeds up the root bridge switch. The link establishment with the non-root bridge switch reduces the interruption time of the multicast service flow and improves the forwarding efficiency.

另外,本发明中的三层交换机可以更换成路由器,不管是三层交换机还是路由器,都属于携带路由选择功能的三层设备,只要它们之间通过包含多物理端口的VLAN相对接,在对接的VLAN接口上运行组播路由协议,在对接的该VLAN包含的所有以太网端口上运行生成树协议,就可以采用本发明的不间断转发的方法,实现本发明的目的。In addition, the layer-3 switch in the present invention can be replaced with a router. No matter it is a layer-3 switch or a router, it is a layer-3 device with a routing function. The multicast routing protocol is run on the VLAN interface, and the spanning tree protocol is run on all the Ethernet ports included in the interconnected VLAN, so that the non-stop forwarding method of the present invention can be adopted to realize the purpose of the present invention.

Claims (8)

1. the method for a multicast data uninterrupted forwarding when the root port of Spanning-Tree Protocol on the nonroot bridge three-layer equipment switches, may further comprise the steps:
101, the root port that the nonroot bridge three-layer equipment is new is set to the multicast forwarding port, described nonroot bridge three-layer equipment is not waited for the transmission cycle arrival of neighbor discovery protocol message, sends neighbor discovery protocol message by new root port to the designated port of root bridge three-layer equipment;
When 102, root bridge three-layer equipment is received described neighbor discovery protocol message, described designated port is set to the multicast forwarding port, described bridge three-layer equipment do not waited for the transmission cycle arrival of neighbor discovery protocol message, responds neighbor discovery protocol message by this designated port to described new root port;
103, after the nonroot bridge three-layer equipment is received the neighbor discovery protocol message of root bridge three-layer equipment response, described new root port and the link establishment between the described designated port, multicast packet continues to transmit by this link.
2. the method for a kind of multicast data uninterrupted forwarding according to claim 1, it is characterized in that, also comprised before step 101: when the root port of Spanning-Tree Protocol on the nonroot bridge three-layer equipment switched, the Spanning-Tree Protocol state of new root port and old root port is respectively to be transmitted and abandons.
3. the method for a kind of multicast data uninterrupted forwarding according to claim 1, it is characterized in that: in the step 101, when the new root port of nonroot bridge three-layer equipment was set to the multicast forwarding port, old root port was set to block the multicast forwarding port.
4. the method for a kind of multicast data uninterrupted forwarding according to claim 1, it is characterized in that: in the step 101, the periodicity neighbor discovery protocol message transmission timer of at first overtime its inside of nonroot bridge three-layer equipment, after described periodicity neighbor discovery protocol message transmission timer is overtime, send neighbor discovery protocol message to the designated port of root bridge three-layer equipment by new root port.
5. the method for a kind of multicast data uninterrupted forwarding according to claim 1, it is characterized in that: in the step 102, the periodicity neighbor discovery protocol message transmission timer of at first overtime its inside of root bridge three-layer equipment, after described periodicity neighbor discovery protocol message transmission timer is overtime, respond neighbor discovery protocol message to described new root port by described designated port.
6. according to the method for the arbitrary described a kind of multicast data uninterrupted forwarding of claim 1 to 5, it is characterized in that: described three-layer equipment is three-tier switch or router.
7. according to the method for the arbitrary described a kind of multicast data uninterrupted forwarding of claim 1 to 5, it is characterized in that: described neighbor discovery protocol message is your good message of protocol-independent multicast or participates in message.
8. the method for a kind of multicast data uninterrupted forwarding according to claim 1, it is characterized in that: described Spanning-Tree Protocol is RSTP or Multiple Spanning Tree Protocol.
CNB2005101020840A 2005-12-01 2005-12-01 A method for uninterrupted forwarding of multicast data Expired - Fee Related CN100417139C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2005101020840A CN100417139C (en) 2005-12-01 2005-12-01 A method for uninterrupted forwarding of multicast data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2005101020840A CN100417139C (en) 2005-12-01 2005-12-01 A method for uninterrupted forwarding of multicast data

Publications (2)

Publication Number Publication Date
CN1859288A CN1859288A (en) 2006-11-08
CN100417139C true CN100417139C (en) 2008-09-03

Family

ID=37298152

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2005101020840A Expired - Fee Related CN100417139C (en) 2005-12-01 2005-12-01 A method for uninterrupted forwarding of multicast data

Country Status (1)

Country Link
CN (1) CN100417139C (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101296105B (en) * 2008-06-11 2011-04-20 杭州华三通信技术有限公司 Multicast fast switching method and system, three-layer network appliance
CN103701703B (en) * 2013-12-31 2017-05-10 新华三技术有限公司 Neighbourhood management method and device
CN108512751B (en) * 2017-02-28 2021-02-23 华为技术有限公司 A port state processing method and network device
CN109818804B (en) * 2019-02-26 2020-04-24 北京邮电大学 Network monitoring method and device
CN113904980B (en) * 2021-09-30 2023-07-18 北京东土科技股份有限公司 A channel switching method, device, switch and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1378365A (en) * 2002-05-10 2002-11-06 北京港湾网络有限公司 IP group broadcast route repeating optimizing method
US20020191628A1 (en) * 2001-06-12 2002-12-19 Actue Communications Corporation Apparatus for multicast forwarding in a virtual local area network environment
US20030079040A1 (en) * 2001-10-19 2003-04-24 Nitin Jain Method and system for intelligently forwarding multicast packets
CN1474563A (en) * 2002-08-06 2004-02-11 ��Ϊ�������޹�˾ Forwarding method of multicast message in network communication
CN1484406A (en) * 2003-08-11 2004-03-24 北京港湾网络有限公司 The Method of Realizing Multicast Accurate to Port Forwarding on Layer 3 Switch

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020191628A1 (en) * 2001-06-12 2002-12-19 Actue Communications Corporation Apparatus for multicast forwarding in a virtual local area network environment
US20030079040A1 (en) * 2001-10-19 2003-04-24 Nitin Jain Method and system for intelligently forwarding multicast packets
CN1378365A (en) * 2002-05-10 2002-11-06 北京港湾网络有限公司 IP group broadcast route repeating optimizing method
CN1474563A (en) * 2002-08-06 2004-02-11 ��Ϊ�������޹�˾ Forwarding method of multicast message in network communication
CN1484406A (en) * 2003-08-11 2004-03-24 北京港湾网络有限公司 The Method of Realizing Multicast Accurate to Port Forwarding on Layer 3 Switch

Also Published As

Publication number Publication date
CN1859288A (en) 2006-11-08

Similar Documents

Publication Publication Date Title
EP3367619B1 (en) Synchronizing multicast state between multi-homed routers in an ethernet virtual private network
CN1980230B (en) Methods for Managing VRRP Groups
CN102638389B (en) The redundancy backup method and system of a kind of TRILL network
EP2533475B1 (en) Method and system for host route reachability in packet transport network access ring
US7269135B2 (en) Methods and systems for providing redundant connectivity across a network using a tunneling protocol
US8498299B2 (en) Flooding-based routing protocol having average-rate and burst-rate control
CN101252427B (en) Switch method of main-standby routers connecting with main-standby interfaces of medium gateways
US9059902B2 (en) Procedures, apparatuses, systems, and computer-readable media for operating primary and backup network elements
CN101394260B (en) Method and apparatus for realizing main standby inverting and load sharing
US7911940B2 (en) Adaptive redundancy protection scheme
WO2011021180A1 (en) Technique for dual homing interconnection between communication networks
JP2008236212A (en) VPN equipment
JP2003008609A (en) Communication device with line redanduncy function
CN102244609A (en) Method for preventing flow interruption caused by failover during access of VPLS (virtual private LAN service) to L3 (network layer 3) and router
CN102957620B (en) Equipment and method for management of MAC (media access control) address table entries in TRILL (transparent interconnection of lots of links) network
CN101800691A (en) Method, equipment and system for establishing data forwarding paths in ethernets
CN101305549A (en) Detect out-of-service links in communication networks
US20140269746A1 (en) Load balancing of logical connections over multi-chassis trunk
CN101087277A (en) A digital home gateway device and its processing method
CN101321077B (en) Reliable multicast method, provider edge upper layer device and system
US20130100854A1 (en) Vpls over multi-chassis trunk
CN105490937A (en) Ethernet virtual network gateway switching method and service provider edge node equipment
WO2013166911A1 (en) Method for processing conflict of identifiers of device groups in network, and route bridge
CN101296126B (en) Link fault announcing method, interface management unit and router
CN100417139C (en) A method for uninterrupted forwarding of multicast data

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
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: 20080903

Termination date: 20191201