WO2008058428A1 - A specific source multicasting method - Google Patents

A specific source multicasting method Download PDF

Info

Publication number
WO2008058428A1
WO2008058428A1 PCT/CN2006/003101 CN2006003101W WO2008058428A1 WO 2008058428 A1 WO2008058428 A1 WO 2008058428A1 CN 2006003101 W CN2006003101 W CN 2006003101W WO 2008058428 A1 WO2008058428 A1 WO 2008058428A1
Authority
WO
WIPO (PCT)
Prior art keywords
port
channel
mode
layer
switch
Prior art date
Application number
PCT/CN2006/003101
Other languages
French (fr)
Chinese (zh)
Inventor
Yancong Ma
Qinxue Li
Original Assignee
Zte Corporation
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 Zte Corporation filed Critical Zte Corporation
Priority to CN200680055256.6A priority Critical patent/CN101480010B/en
Priority to PCT/CN2006/003101 priority patent/WO2008058428A1/en
Publication of WO2008058428A1 publication Critical patent/WO2008058428A1/en

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]
    • H04L12/46Interconnection of networks
    • H04L12/4641Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/02Details
    • H04L12/16Arrangements for providing special services to substations
    • H04L12/18Arrangements for providing special services to substations for broadcast or conference, e.g. multicast
    • H04L12/185Arrangements for providing special services to substations for broadcast or conference, e.g. multicast with management of multicast group membership
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/16Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]

Definitions

  • the present invention relates to the field of communications, and in particular to a specific source multicast method, which relates to
  • IGMPv3 Internet Group Member Protocol version 3
  • IGMPvl and IGMPv2 Internet Multicast Management Protocol Version 1 and Version 2
  • ASM Any Source Multicast
  • any source multicast system there are some inherent defects, such as address allocation and access control issues, which cannot prevent address conflicts between multiple applications.
  • Once a receiver joins a multicast group it will forward any source to the multicast group. The data. Even if a multicast source obtains a multicast group address to send data, there is no guarantee that no other source will use this address.
  • SSM Specific Source Multicast
  • the service provided by the SSM to the network layer is a "channel” that uses the destination IP address of the specific multicast (G) and the source IP address.
  • the address S (Source Address) is "channel", and the recipient can receive these datagrams by joining the (S, G) channel.
  • SSM mitigates and compensates for the shortcomings of ASM.
  • the specific source multicast defines the IP address G and the source IP address S (S, G), and the G is extended by S, for example (SI, G) and (S2, G), although they have the same The group address G, but represents two channels by S1 and S2, respectively. This avoids the problem of globally assigning multicast addresses of specific source multicast destinations.
  • Each multicast source can separately handle address conflicts between multiple channels it creates.
  • access control when a receiver joins a (S, G) channel, it can only receive data sent by the source S. Instead, any host can send data to an ASM host multicast group.
  • the specific source multicast is automatically guaranteed not to There are other senders that send data on one channel. This makes it much more difficult to send spam on a particular source multicast channel than on any source multicast group.
  • the SSM In the process of processing a well-known multicast source, the SSM only needs to maintain a source-based forwarding tree. The shared tree system in the multicast routing protocol is no longer needed. This will greatly reduce the multicast routing structure.
  • Layer 3 routers can control the forwarding of multicast packets.
  • multicast packets inevitably pass through some Layer 2 switching devices, such as Layer 2 switches, especially in a LAN environment. If a Layer 2 switch does not process multicast packets, it floods all interfaces of the Layer 2 switch, causing waste of resources and affecting performance.
  • IGMP Snooping is usually used to solve this problem in any source multicast system.
  • the receiving host sends an IGMP member report message.
  • the Layer 2 switch passes the Layer 2 switch, the Layer 2 switch listens to the message and records the relationship between the group member and the interface.
  • the multicast packet When a multicast packet is received, the multicast packet is forwarded only to the interface that has the member of the group. After the Layer 2 switch processes the IGMP protocol packet, it forwards the protocol packet to the Layer 3 device interface intact.
  • the IGMP multicast member management mechanism is designed for the Layer 3, and the Layer 3 router can control the forwarding of multicast packets.
  • Layer 2 switches in a specific source multicast cannot process multicast packets, causing flooding to all interfaces of the Layer 2 switch, resulting in waste of resources and impact on performance. .
  • the present invention is directed to a specific source multicast method for solving the problem that a specific source multicast cannot be implemented on a layer 2 switch in the related art.
  • a specific source multicast method for a network including a Layer 2 switch and a Layer 3 switch comprising the steps of: dividing a VLAN resource into a specific source multicast dedicated in a Layer 2 switch, and The working area of the VLAN resource is divided; the (S, G) channel is added in the working domain; the IGMP protocol packet sent by the port is matched with the (S, G) channel in the VLAN resource; and the packet is reassembled according to the matching situation. And forwarded to the Layer 3 switch.
  • adding the (S, G) channel in the working domain includes the following steps: Specifying a source port and a receiving port in the working domain, where the source port is a program source port in the VLAN resource, and the receiving port Is the section view port in the VLAN resource.
  • the matching includes the following steps: If the message is consistent with the (S, G) channel, the port is added to the (S, G) channel; if the message conflicts with the (S, G) channel, Then delete the port in the (S, G) channel.
  • determining whether the packet and the (S, G) channel are consistent includes the following steps: parsing all channel request information in the packet, and then (S, G) channel configured on the layer 2 switch Matches to determine whether the port is a traffic port or a traffic-free port. If it is a traffic port, it is the same, otherwise it is a conflict.
  • parsing all the channel request information in the packet includes the following steps: Decomposing the packet in units of group records to obtain channel request information, where each channel of the request is configured as a request type ⁇ R , (S, G) ⁇ , mark the request type as ⁇ Join, (S, G) ⁇ and ⁇ Leave, (S, G) ⁇ .
  • the judgment criterion of ⁇ Join, (S, G) ⁇ is as follows: If the record type is MODE-IS-INCLUDE or CHANGE-TO-INCLUDE-MODE, the message contains (S, G) If the record type is MODE—IS—EXCLUDE or CHANGE—TO—EXCLUDE—MODE, the message does not contain (S, G) items; the record type is ALLOW_NEW—SOURCES, and the current filter mode is INCLUDE, then the message contains ( S, G); The record type is BLOCK_OLD_SOURCES, and the current filter mode is EXCLUDE, then the message contains (S, G) entries.
  • the judgment criterion of ⁇ Leave, (S, G) ⁇ is as follows: If the record type is MODE JS—INCLUDE or CHANGE_TO—INCLUDE_MODE, the message does not contain (S, G) Item; record type is MODE-IS-EXCLUDE or CHANGE-TO-EXCLUDE-MODE, the message does not contain (S, G) items; record type is ALLOW_NEW__SOURCES, the current filter mode is EXCLUDE, then the message contains (S, G); the record type is BLOCK_OLD_SOURCES, and the current filter mode is INCLUDE, then the message contains (S, G) entries.
  • adding a port includes the following steps: When the channel request information of the port is ⁇ Join, (S, G) ⁇ , the (S, G) channel is searched, if it does not match, the ⁇ Join, (S, G ) ⁇ will be discarded; if it matches, add a port in the (S, G) channel.
  • deleting a port includes the following steps: When the channel request information of the port is ⁇ Leave, (S, G) ⁇ , the port is deleted in the (S, G) channel.
  • reassembling the IGMP protocol packet and forwarding it to the Layer 3 switch includes the following steps:
  • the Layer 2 switch encapsulates the IGMP protocol packet into multiple MODE-ISJNCLUDE records [G, SI], [G, S2], [G, S3] . ⁇ .., and then forward these records to the Layer 3 switch through different channels.
  • the present invention achieves the following technical effects: This document provides a method for implementing a Layer 2 switch IGMPv3, which is a "Dominated IGMPv3 Snooping" method.
  • the implementation method performs effective analysis and processing on IGMP v3 packets based on the multicast deployment plan of the Layer 2 and Layer 3 devices, and then splits, converts, records, and transmits the packets to the upper-layer routers.
  • This method is no longer a simple monitoring and recording. It needs to monitor and process IGMPv3 packets according to the deployment of multicast traffic.
  • the method effectively implements the support of the Layer 2 switch for IGMPv3, so that the Layer 2 switch can have a strong specific source multicast function under the existing conditions.
  • FIG. 1 shows a flow chart of a specific source multicast method according to the present invention
  • FIG. 2 shows a flow chart of a specific source multicast method according to an embodiment of the present invention.
  • the present invention provides a specific source multicast implementation method.
  • Step S102 Divide a VLAN resource into a specific source multicast in a Layer 2 switch, and divide the VLAN resource. a working domain; step S104, adding a (&, G) channel in the working domain; step S106, matching an IGMP protocol packet sent by the port with the (S, G) channel in the VLAN resource; and step S108, according to the matching In the case, the message is reassembled and forwarded to the Layer 3 switch.
  • Step S104 may include the following steps: Specifying a source port and a receiving port in the working domain, where the source port is a program source port in the VLAN resource, and the receiving port is a program viewing port in the VLAN resource.
  • the matching may include the following steps: If the message is consistent with the (S, G) channel, then the port is added to the (S, G) channel; if the message and the (S, G) channel conflict, then in the (S, G) channel Delete the port.
  • Determining whether the packet is consistent with the (S, G) channel may include the following steps: parsing all channel request information in the packet, and then matching with the (S, G) channel configured on the layer 2 switch, thereby determining that the port is The flow port is still a no-flow port. If it is a flow-through port, it is the same, otherwise it is a conflict.
  • Parsing all the channel request information in the message may include the following steps: Decomposing the message in units of group records to obtain channel request information, wherein each channel of the request is configured as a request type ⁇ R, (S, G) ⁇ , Request types are labeled ⁇ Join, (S, G) ⁇ and ⁇ Leave, (S, G) ⁇ .
  • the judgment criterion of ⁇ Join, (S, G) ⁇ can be as follows: Record type is MODE—IS—INCLUDE or CHANGE—TO—INCLUDE—MODE, then the message contains (S, G); Record type is MODE—IS— EXCLUDE or CHANGE_TO_EXCLUDE_MODE, the message does not contain (S, G) items; the record type is ALLOW_NEW_SOURCES, the current filter mode is INCLUDE, then the message contains (S, G) items; The type is BLOCK_OLD_SOURCES, and the current filtering mode is EXCLUDE, then the message contains (S, G) items.
  • the judgment criterion of ⁇ Leave, (S, G) ⁇ can be as follows: If the record type is MODE-IS-INCLUDE or CHANGE_TO-INCLUDE-MODE, the message does not contain (S, G) items; the record type is MODE-IS — EXCLUDE or CHANGE— TO— EXCLUDE— MODE , the message does not contain ( S, G) items; the record type is ALLOW_NEW_SOURCES, the current filter mode is EXCLUDE, Then the message contains (S, G) items; the record type is BLOCK_OLD_SOURCES, and the current filtering mode is INCLUDE, then the message contains (S, G) items.
  • Adding a port may include the following steps: When the channel request information of the port is ⁇ Join, (S, G) ⁇ , the (S, G) channel is searched, and if it does not match, the ⁇ Join, (S, G) ⁇ will be thrown. Discard; if matching, add a port in the (S, G) channel. Deleting a port can include the following steps: When the channel request information of the port is ⁇ Leave, (S, G) ⁇ , the port is deleted in the (S, G) channel.
  • Reassembling IGMP protocol packets and forwarding them to Layer 3 switches can include the following steps: Layer 2 switches encapsulate IGMP protocol packets into multiple MODE-IS-INCLUDE records [G, SI], [G, S2], [G, S3 ] , . . . . . , then forward these records to the Layer 3 switch through different channels.
  • the present invention will be described in detail below with reference to FIG. 2 and in conjunction with the embodiments.
  • the method includes the following steps: Step S202. Divide some VLAN (Virtual Local Area Network) resources into multicast for the Layer 2 device. And divide the working domain of the multicast VLAN. Step S204. In the working domain of the multicast VLAN, add several (S, G) channels. Step S206.
  • step S208 Match the IGMP protocol packet sent by the port with the (S, G) channel in the multicast VLAN.
  • step S208 if they are consistent, the port is added to the (S, G) channel.
  • step S210 if there is a conflict, the port is deleted in the (S, G) channel.
  • Step S212 In the multicast VLAN, the packet is reassembled to the Layer 3 device according to the matching with the (S, G) channel.
  • a Layer 2 device identifies a data stream and exchanges data through a VLAN ID (VID) and a destination MAC (Destination Media Access Control) address (DA) in data exchange.
  • VIP VLAN ID
  • DA Destination Media Access Control
  • the present invention divides two parts of VLAN resources on the Layer 2 switch, and a part of the multicast data stream uses a multicast VLAN, and the Layer 3 switching device connected to the Layer 2 switch passes the multicast stream through the multicast VLAN. It is sent to the uplink port of the Layer 2 switch; another part of the VLAN is given to the (S, G) channel.
  • Each (S, G) channel corresponds to an SGVLAN, and the SGVALN is assigned to the (S, G) channel of a specific source address and group address through the network management configuration.
  • the present invention corresponds to the (S, G) channel multicast stream and the layer 2 forwarding stream (VID, DA) by configuring the (S, G) channel.
  • each multicast VLAN corresponds to the same VLAN interface of the Layer 3 device.
  • the (S, G) channel on the Layer 2 switch belongs to a multicast VLAN.
  • IGMPv3 report packets and query packets are processed during IGMPv3 packet processing.
  • the terminal host connected to the Layer 2 switch will send a report message to indicate the (S, G) channel that you want to join or leave.
  • a complete group record mainly consists of three parts, group record type, one group address (G), and one column source address (SI, S2, S3).
  • the types of IGMPv3 group records are as follows:
  • the current state is the inclusion mode (MODE_IS JNCLUDE)
  • the current state is the exclusion mode (MODE_IS—EXCLUDE)
  • Source address list transformation allowing new source addresses ( ALLOW - NEW - SOURCES )
  • Source address list conversion masking the old source address (BLOCK_OLD_SOURCES)
  • the group address and the source address list correspond to the (S, G) channel's permissions for the port are different, whether it is What combination, for the (S, G) channel on the Layer 2 switch, only two results are produced.
  • One is that the port allows the traffic in the (S, G) channel to be received, for the (S, G) channel.
  • the traffic port; the second is that the port is prohibited from receiving traffic in the (S, G) channel, which is the forbidden port of the (S, G) channel.
  • the core problem of the Layer 2 switch in the processing of IGMPv3 report packets is to resolve all (S, G) channels in the packets and then match the (S, G) channels configured on the Layer 2 switch.
  • the port is a flow port or a no-flow port, if it is a flow port, it needs to be at (S, G) The port is added to the channel, otherwise the port is deleted.
  • the Layer 2 switch After processing the channel according to the report message, the Layer 2 switch reassembles the message and encapsulates the message record [G, SI, S2, S3...] into several records [G, SI ], [G, S2], [G, S3], etc., then forward these records to the third layer through different multicast VLANs according to the membership of the (S, G) channel in the Layer 2 switch in the multicast VLAN. device. Specifically, when the Layer 2 switch receives the multicast address as G from the host, and the source address list is MODE_IS-INCLUDE of S1, S2, and S3, for example, the Layer 2 switch receives an IGMP report message from the host.
  • record type is MODE-ISJNCLUDE
  • multicast address is G
  • source address list is SI
  • the port will be decomposed into three MODE-IS-INCLUDE, (SI, G) records will be sent to the multicast VLAN1 interface, (S2, G) records will be sent to the multicast VLAN2 interface, (S3, G) records will be Send to the multicast VLAN3 interface.
  • Such (SI, G), (S2, G), (S3, G) multicast streams will be sent to the Layer 2 switch through different VLANs 1, 2, 3, for the same group address data stream, VLAN1, VLAN2, VLAN3 Corresponding to different VLAN IDs, so that different source IDs can be distinguished by different VLAN IDs, and the correspondence between (S, G) and (VID, G) is realized.
  • the source port and the receiving port are first specified in the multicast working domain, where the source port is a program source port in the multicast VLAN, and for the layer 2 switch, it is usually a port associated with the layer 3 router; It is the program viewing port, usually the program out port, so the packets of the non-source port and the receiving port will be ignored. Then, the report message is decomposed in units of group records, and the request information of the port for several channels is obtained.
  • Each channel is configured as a request type (R: Request) plus channel information (S, G), abbreviated as ⁇ R, (S , G) ⁇
  • the request type has two types: Join and Leave, which are labeled ⁇ Join, (S, G) W ⁇ Leave, (S, G) ⁇ .
  • the judgment criteria of ⁇ Join, (S, G) ⁇ in the message are as follows:
  • the record type is MODE—IS—INCLUDE or CHANGE—TO—INCLUDE—MODE, and the message contains (S, G) items;
  • the record type is MODE_IS—EXCLUDE or CHANGEJTO—EXCLUDE—MODE, and the message does not contain (S, G) items; 3.
  • the record type is ALLOW-NEW-SOURCES, the current filter mode is INCLUDE, and the message contains (S, G) items; _ ;
  • the record type is BLOCK_OLD_SOURCES, and the current filter mode is EXCLUDE, the message contains (S, G) items; the judgment criteria of ⁇ Leave, (S, G) ⁇ in the message are as follows:
  • the record type is MODE—ISJNCLUDE or CHANGE—TOJNCLUDE—MODE, and the message does not contain (S, G) items;
  • the type of record is MODE—IS—EXCLUDE or CHANGE—TO—EXCLUDE—MODE, and the message does not contain (S, G) items;
  • the record type is ALLOW-NEW_SOURCES, the current filter mode is EXCLUDE, and the message contains (S, G) entries;
  • Step 1 Decompose the IGMPv3 report received by the receiving port in units of group records. , get a series of ⁇ Join, (S, G) p ⁇ Leave, (S, G) ⁇ , and then process them one by one.
  • Step 2. When ⁇ Join, (S, G) ⁇ is processed, search for the (S, G) channel configured in each multicast VLAN configured on the Layer 2 switch.
  • the received query messages will be sent to all receiving ports in the multicast VLAN.
  • the present invention achieves the following beneficial effects.
  • This document provides a method for implementing a Layer 2 switch IGMPv3, which is a "Dominated IGMPv3 Snooping" method.
  • the implementation method effectively analyzes and processes IGMP v3 packets based on the multicast deployment plan of the Layer 2 and Layer 3 devices, splits, converts, records, and delivers them to the upper-layer routers. This method is no longer a simple monitoring and recording. It needs to monitor and process IGMPv3 packets according to the deployment of multicast traffic.
  • the method effectively implements the support of the Layer 2 switch for IGMPv3, so that the Layer 2 switch can have a strong specific source multicast function in the existing device.

Abstract

A specific source multicasting method applied in the network comprising a layer 2 switch and a layer 3 switch, includes the following steps: dividing the VLAN resource specially for the specific source multicast in the layer 2 switch and dividing the work domain of the VLAN resource; adding the (S,G) channel in the work domain; matching the IGMP protocol message sent from the port with the (S,G) channel in the VLAN resource; and according to the match, reforming the message and forwarding the message to the layer 3 switch.

Description

特定源组播方法 技术领域 本发明涉及通信领域, 具体而言, 涉及一种特定源组播方法, 其涉及 TECHNICAL FIELD The present invention relates to the field of communications, and in particular to a specific source multicast method, which relates to
IGMPv3 ( Internet Group Member Protocol version 3 )在二层交换机设备上的实现。 背景技术 组播技术实现了从单点到多点的高效传输, 相对于单播和广播来说有效地 节省了带宽资源。 利用组播功能可以方便地提供增值业务, 如网络电视、 在线直 播、 远程教育、 远程医疗、 网络电台、 实时视频会议等业务。 互联网组播管理协议版本 1和版本 2 ( IGMPvl和 IGMPv2 ) 中定义的组播 模型都是任意源组播(Any Source Multicast, 缩写为 ASM )模型, 这个模型支 持任意多个发送者, 任意一个终端主机(或路由器)都可以发送数据给一个主机 组, 即使它可能不是该组成员。 在任意源组播体系下, 存在一些天生缺陷, 如地 址分配和访问控制问题, 无法防止多个应用之间地址冲突,接收者一旦加入一个 组播组, 将会转发任意源发送到组播组的数据。 即使一个组播源获得了一个组播 组地址来发送数据, 也不能保证没有其他源会使用这个地址。 IGMPv3 (Internet Group Member Protocol version 3) is implemented on a Layer 2 switch device. BACKGROUND OF THE INVENTION Multicast technology enables efficient transmission from a single point to multiple points, effectively saving bandwidth resources compared to unicast and broadcast. Multicast functions can be easily used to provide value-added services such as Internet TV, online direct broadcast, distance education, telemedicine, Internet radio, and real-time video conferencing. The multicast models defined in Internet Multicast Management Protocol Version 1 and Version 2 (IGMPvl and IGMPv2) are all Any Source Multicast (ASM) models. This model supports any number of senders, any one of the terminals. The host (or router) can send data to a host group, even though it may not be a member of the group. In any source multicast system, there are some inherent defects, such as address allocation and access control issues, which cannot prevent address conflicts between multiple applications. Once a receiver joins a multicast group, it will forward any source to the multicast group. The data. Even if a multicast source obtains a multicast group address to send data, there is no guarantee that no other source will use this address.
IGMPv3中定义的是特定源组播( Specific Source Multicast, 缩写为 SSM ) 模型, SSM提供给网络层的服务是一个 "通道", 它用特定组播的目的 IP地址 G ( Group Address )和源 IP地址 S ( Source Address )进行"通道", 接受者只要加 入该 (S, G )通道可以收到这些数据报。 SSM緩解和弥补了 ASM的缺陷。 在地址分配上, 特定源组播定义了 IP地址 G和源 IP地址 S ( S , G ) , 通 过 S对 G进行了扩展, 例如(SI , G )和(S2, G ) , 虽然拥有相同的组地址 G, 但是通过 S1和 S2来分別代表两个通道。这样就避开了全局分配特定源组播目的 组播地址的问题,每一个组播源能够单独处理它所创建的多个通道之间的地址冲 突问题。 在访问控制上, 当接收者加入到一个( S , G )通道, 它只能接收到源 S发 送的数据。 相反, 任何主机都可以向一个 ASM主机组播组发送数据。 在一个发 送者选择了一个 ( S , G )通道来发送数据的同时, 特定源組播自动地保证不会 有其他发送者在间一个通道上发送数据。这使得在特定源組播通道上发送垃圾信 息比在任意源组播组上发送要困难得多。 在处理众所周知的组播源过程中, SSM只需要简单的维护一棵基于源的转 发树, 而組播路由协议中的共享树体系已经不再需要, 这样一方面将大大降低组 播路由结构的复杂性, 另一方面可以直接部署组播网络。 三层路由器可以对组播报文的转发进行控制。 但大多数情况下, 组播报文 不可避免地要经过一些二层交换设备, 如二层交换机, 尤其是在局域网环境里。 如果二层交换机不对组播报文进行处理, 就会洪泛给二层交换机的所有接口,造 成资源浪费和影响性能使用。 在任意源组播体系中通常采用 IGMP监听( IGMP Snooping )解决这个问题。 任意源组播在 IGMP监听工作过程中,接收主机发出 IGMP成员报告消息, 在经 过二层交换机时,二层交换机对这个消息进行监听并记录下来, 形成组成员和接 口的对应关系, 二层交换机在收到组播数据报文时,根据组成员和接口的对应关 系, 仅向具有组成员的接口转发组播报文。 二层交换机处理完 IGMP协议包后, 将向三层设备接口原封不动的转发该协议包。 然而在根据 IGMPv3 的特定源组播中, IGMP组播成员管理机制是针对第 三层设计的, 三层路由器可以对组播报文的转发进行控制。 对于二层交换机, 尤 其是在局域网环境里, 特定源組播中的二层交换机不能够对组播报文进行处理, 导致洪泛给二层交换机的所有接口, 从而造成资源浪费和影响性能使用。 因此, 人们需要一种在包括二层交换机的网络上实现特定源组播的解决方 案, 能够解决上述相关技术中的问题。 发明内容 本发明旨在提供一种特定源组播方法, 用于解决上述相关技术中二层交换 机上无法实现特定源组播等问题。 根据本发明的一个方面, 提供了一种用于包括二层交换机和三层交换机的 网络的特定源组播方法, 包括以下步驟: 在二层交换机中划分 VLAN资源给特 定源組播专用, 并且划分 VLAN资源的工作域; 在工作域中增加(S, G )通道; 在 VLAN资源中对端口发送的 IGMP协议报文和 ( S , G )通道进行匹配; 以及 根据匹配的情况, 重组报文并转发给三层交换机。 在上述的特定源组播方法中, 在工作域中增加(S, G)通道包括以下步骤: 在工作域中规定源端口和接收端口, 其中源端口为 VLAN 资源中的节目来源端 口, 接收端口是 VLAN资源中的节 观看端口。 在上述的特定源组播方法中, 匹配包括以下步驟: 假如报文和(S, G)通 道一致, 那么将端口加入(S, G)通道; 假如报文和 (S, G)通道冲突, 则在 (S, G)通道中删除端口。 在上述的特定源組播方法中, 判断报文和(S, G)通道是否一致包括以下 步骤: 解析报文中所有的通道请求信息, 然后和二层交换机上配置的 (S, G) 通道进行匹配, 以此判断端口是通流端口还是禁流端口,假如是通流端口则是一 致, 否则为沖突。 在上述的特定源组播方法中, 解析报文中所有的通道请求信息包括以下步 骤: 以组记录为单位分解报文得到通道请求信息, 其中, 将请求的每一个通道构 成为请求类型 {R, (S, G) }, 将请求类型分別标为 {Join, (S, G) }和 {Leave, (S, G) }。 在上述的特定源组播方法中, {Join, (S, G) }的判断准则如下: 记录类 型为 MODE— IS— INCLUDE或者 CHANGE— TO— INCLUDE— MODE, 则报文包含 ( S , G ) 项 ; 记 录 类 型 为 MODE— IS— EXCLUDE 或 者 CHANGE— TO— EXCLUDE— MODE , 则报文中不包含( S , G ) 项; 记录类型为 ALLOW_NEW— SOURCES, 当前过滤模式是 INCLUDE, 则报文包含(S, G) 项; 记录类型为 BLOCK— OLD— SOURCES, 并且当前过滤模式是 EXCLUDE, 则 报文包含(S, G) 项。 在上述的特定源组播方法中, {Leave, (S, G) }的判断准则如下: 记录类 型为 MODE JS— INCLUDE或者 CHANGE— TO— INCLUDE— MODE, 则报文不包 含 ( S , G ) 项 ; 记 录 类 型 为 MODE— IS— EXCLUDE 或 者 CHANGE一 TO— EXCLUDE— MODE, 则报文中不包含( S , G ) 项; 记录类型为 ALLOW_NEW__SOURCES, 当前过滤模式是 EXCLUDE , 则报文包含( S , G) 项; 记录类型为 BLOCK— OLD— SOURCES, 并且当前过滤模式是 INCLUDE, 则 报文包含(S, G) 项。 在上述的特定源组播方法中, 增加端口包括以下步骤: 当端口的通道请求 信息为 {Join, (S, G) }时, 搜索 (S, G)通道, 假如不匹配, 该 {Join, (S, G ) }将被扔弃; 假如匹配, 在(S, G )通道中增加端口。 在上述的特定源组播方法中, 删除端口包括以下步骤: 当端口的通道请求 信息为 {Leave, ( S, G ) }时, 在(S, G )通道中删除端口。 在上述的特定源組播方法中, 重組 IGMP协议报文并转发给三层交换机包 括以下步骤: 二层交换机将 IGMP协议报文封装成多条 MODE— ISJNCLUDE记 录 [G, SI] , [G, S2] , [G, S3] .·· .·. , 然后通过不同的通道, 将这些记录转发给 三层交换机。 通过上述技术方案, 本发明实现了如下技术效果: 本文提供一种二层交换机 IGMPv3 的实现方法, 是一种 "受控监听 ( Dominated IGMPv3 Snooping )"方法。该实现方法在任意源组播 IGMP Snooping 的基础上, 根据二三层设备組播部署规划, 对 IGMP v3报文进行有效的分析处 理, 进行拆分, 转换, 记录, 再传递给上层路由器。 该方法不再是单纯的监听和 记录, 它需要根据组播流量的部署情况对 IGMPv3报文进行监听和处理。 该方法 有效地实现了二层交换机对 IGMPv3的支持,使二层交换机在现有条件就可以具 备较强的特定源组播功能。 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示出了根据本发明的特定源組播方法的流程图; 图 2示出了根据本发明实施例的特定源组播方法的流程图。 具体实施方式 本发明提供了一种特定源组播实现方法, 如图 1所示, 其包括以下步骤: 步骤 S102, 在二层交换机中划分 VLAN资源给特定源组播专用, 并且划分 VLAN资源的工作域; 步骤 S104, 在工作域中增加 (&, G)通道; 步骤 S106, 在 VLAN资源中对端口发送的 IGMP协议报文和(S, G)通 道进行匹配; 以及 步骤 S 108, 根据匹配的情况, 重组报文并转发给三层交换机。 步骤 S104可包括以下步骤: 在工作域中规定源端口和接收端口, 其中源端 口为 VLAN资源中的节目来源端口,接收端口是 VLAN资源中的节目观看端口。 匹配可包括以下步骤: 假如报文和(S, G)通道一致, 那么将端口加入(S, G)通道; 假如报文和 (S, G)通道冲突, 则在(S, G)通道中删除端口。 判断报文和(S, G)通道是否一致可包括以下步骤: 解析报文中所有的通 道请求信息, 然后和二层交换机上配置的 (S, G)通道进行匹配, 以此判断端 口是通流端口还是禁流端口, 假如是通流端口则是一致, 否则为冲突。 解析报文中所有的通道请求信息可包括以下步骤: 以組记录为单位分解报 文得到通道请求信息,其中,将请求的每一个通道构成为请求类型 {R, (S, G)}, 将请求类型分别标为 {Join, (S, G) }和 {Leave, (S, G) }。 Defined in IGMPv3 is the Specific Source Multicast (SSM) model. The service provided by the SSM to the network layer is a "channel" that uses the destination IP address of the specific multicast (G) and the source IP address. The address S (Source Address) is "channel", and the recipient can receive these datagrams by joining the (S, G) channel. SSM mitigates and compensates for the shortcomings of ASM. In the address allocation, the specific source multicast defines the IP address G and the source IP address S (S, G), and the G is extended by S, for example (SI, G) and (S2, G), although they have the same The group address G, but represents two channels by S1 and S2, respectively. This avoids the problem of globally assigning multicast addresses of specific source multicast destinations. Each multicast source can separately handle address conflicts between multiple channels it creates. In access control, when a receiver joins a (S, G) channel, it can only receive data sent by the source S. Instead, any host can send data to an ASM host multicast group. When a sender selects an (S, G) channel to send data, the specific source multicast is automatically guaranteed not to There are other senders that send data on one channel. This makes it much more difficult to send spam on a particular source multicast channel than on any source multicast group. In the process of processing a well-known multicast source, the SSM only needs to maintain a source-based forwarding tree. The shared tree system in the multicast routing protocol is no longer needed. This will greatly reduce the multicast routing structure. Complexity, on the other hand, can directly deploy a multicast network. Layer 3 routers can control the forwarding of multicast packets. In most cases, multicast packets inevitably pass through some Layer 2 switching devices, such as Layer 2 switches, especially in a LAN environment. If a Layer 2 switch does not process multicast packets, it floods all interfaces of the Layer 2 switch, causing waste of resources and affecting performance. IGMP Snooping is usually used to solve this problem in any source multicast system. During the IGMP snooping process, the receiving host sends an IGMP member report message. When the Layer 2 switch passes the Layer 2 switch, the Layer 2 switch listens to the message and records the relationship between the group member and the interface. When a multicast packet is received, the multicast packet is forwarded only to the interface that has the member of the group. After the Layer 2 switch processes the IGMP protocol packet, it forwards the protocol packet to the Layer 3 device interface intact. However, in the specific source multicast according to IGMPv3, the IGMP multicast member management mechanism is designed for the Layer 3, and the Layer 3 router can control the forwarding of multicast packets. For a Layer 2 switch, especially in a LAN environment, Layer 2 switches in a specific source multicast cannot process multicast packets, causing flooding to all interfaces of the Layer 2 switch, resulting in waste of resources and impact on performance. . Therefore, there is a need for a solution for implementing a specific source multicast on a network including a Layer 2 switch, which can solve the problems in the related art described above. SUMMARY OF THE INVENTION The present invention is directed to a specific source multicast method for solving the problem that a specific source multicast cannot be implemented on a layer 2 switch in the related art. According to an aspect of the present invention, a specific source multicast method for a network including a Layer 2 switch and a Layer 3 switch is provided, comprising the steps of: dividing a VLAN resource into a specific source multicast dedicated in a Layer 2 switch, and The working area of the VLAN resource is divided; the (S, G) channel is added in the working domain; the IGMP protocol packet sent by the port is matched with the (S, G) channel in the VLAN resource; and the packet is reassembled according to the matching situation. And forwarded to the Layer 3 switch. In the specific source multicast method described above, adding the (S, G) channel in the working domain includes the following steps: Specifying a source port and a receiving port in the working domain, where the source port is a program source port in the VLAN resource, and the receiving port Is the section view port in the VLAN resource. In the specific source multicast method described above, the matching includes the following steps: If the message is consistent with the (S, G) channel, the port is added to the (S, G) channel; if the message conflicts with the (S, G) channel, Then delete the port in the (S, G) channel. In the foregoing specific source multicast method, determining whether the packet and the (S, G) channel are consistent includes the following steps: parsing all channel request information in the packet, and then (S, G) channel configured on the layer 2 switch Matches to determine whether the port is a traffic port or a traffic-free port. If it is a traffic port, it is the same, otherwise it is a conflict. In the specific source multicast method described above, parsing all the channel request information in the packet includes the following steps: Decomposing the packet in units of group records to obtain channel request information, where each channel of the request is configured as a request type {R , (S, G) }, mark the request type as {Join, (S, G) } and {Leave, (S, G) }. In the above specific source multicast method, the judgment criterion of {Join, (S, G) } is as follows: If the record type is MODE-IS-INCLUDE or CHANGE-TO-INCLUDE-MODE, the message contains (S, G) If the record type is MODE—IS—EXCLUDE or CHANGE—TO—EXCLUDE—MODE, the message does not contain (S, G) items; the record type is ALLOW_NEW—SOURCES, and the current filter mode is INCLUDE, then the message contains ( S, G); The record type is BLOCK_OLD_SOURCES, and the current filter mode is EXCLUDE, then the message contains (S, G) entries. In the above specific source multicast method, the judgment criterion of {Leave, (S, G) } is as follows: If the record type is MODE JS—INCLUDE or CHANGE_TO—INCLUDE_MODE, the message does not contain (S, G) Item; record type is MODE-IS-EXCLUDE or CHANGE-TO-EXCLUDE-MODE, the message does not contain (S, G) items; record type is ALLOW_NEW__SOURCES, the current filter mode is EXCLUDE, then the message contains (S, G); the record type is BLOCK_OLD_SOURCES, and the current filter mode is INCLUDE, then the message contains (S, G) entries. In the specific source multicast method described above, adding a port includes the following steps: When the channel request information of the port is {Join, (S, G) }, the (S, G) channel is searched, if it does not match, the {Join, (S, G ) } will be discarded; if it matches, add a port in the (S, G) channel. In the specific source multicast method described above, deleting a port includes the following steps: When the channel request information of the port is {Leave, (S, G) }, the port is deleted in the (S, G) channel. In the specific source multicast method described above, reassembling the IGMP protocol packet and forwarding it to the Layer 3 switch includes the following steps: The Layer 2 switch encapsulates the IGMP protocol packet into multiple MODE-ISJNCLUDE records [G, SI], [G, S2], [G, S3] .·· .., and then forward these records to the Layer 3 switch through different channels. Through the above technical solution, the present invention achieves the following technical effects: This document provides a method for implementing a Layer 2 switch IGMPv3, which is a "Dominated IGMPv3 Snooping" method. The implementation method performs effective analysis and processing on IGMP v3 packets based on the multicast deployment plan of the Layer 2 and Layer 3 devices, and then splits, converts, records, and transmits the packets to the upper-layer routers. This method is no longer a simple monitoring and recording. It needs to monitor and process IGMPv3 packets according to the deployment of multicast traffic. The method effectively implements the support of the Layer 2 switch for IGMPv3, so that the Layer 2 switch can have a strong specific source multicast function under the existing conditions. Other features and advantages of the invention will be set forth in the description which follows, and The objectives and other advantages of the invention will be realized and attained by the <RTI BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the drawings: FIG. 1 shows a flow chart of a specific source multicast method according to the present invention; FIG. 2 shows a flow chart of a specific source multicast method according to an embodiment of the present invention. The present invention provides a specific source multicast implementation method. As shown in FIG. 1 , the method includes the following steps: Step S102: Divide a VLAN resource into a specific source multicast in a Layer 2 switch, and divide the VLAN resource. a working domain; step S104, adding a (&, G) channel in the working domain; step S106, matching an IGMP protocol packet sent by the port with the (S, G) channel in the VLAN resource; and step S108, according to the matching In the case, the message is reassembled and forwarded to the Layer 3 switch. Step S104 may include the following steps: Specifying a source port and a receiving port in the working domain, where the source port is a program source port in the VLAN resource, and the receiving port is a program viewing port in the VLAN resource. The matching may include the following steps: If the message is consistent with the (S, G) channel, then the port is added to the (S, G) channel; if the message and the (S, G) channel conflict, then in the (S, G) channel Delete the port. Determining whether the packet is consistent with the (S, G) channel may include the following steps: parsing all channel request information in the packet, and then matching with the (S, G) channel configured on the layer 2 switch, thereby determining that the port is The flow port is still a no-flow port. If it is a flow-through port, it is the same, otherwise it is a conflict. Parsing all the channel request information in the message may include the following steps: Decomposing the message in units of group records to obtain channel request information, wherein each channel of the request is configured as a request type {R, (S, G)}, Request types are labeled {Join, (S, G) } and {Leave, (S, G) }.
{Join, (S, G) }的判断准则可如下: 记录类型为 MODE— IS— INCLUDE或 者 CHANGE— TO— INCLUDE— MODE , 则报文包含 (S, G) 项; 记录类型为 MODE— IS— EXCLUDE或者 CHANGE— TO— EXCLUDE— MODE, 则报文中不包含 (S, G)项;记录类型为 ALLOW— NEW— SOURCES,当前过滤模式是 INCLUDE, 则报文包含(S, G)项; 记录类型为 BLOCK—OLD— SOURCES, 并且当前过滤 模式是 EXCLUDE, 则报文包含(S, G)项。 The judgment criterion of {Join, (S, G) } can be as follows: Record type is MODE—IS—INCLUDE or CHANGE—TO—INCLUDE—MODE, then the message contains (S, G); Record type is MODE—IS— EXCLUDE or CHANGE_TO_EXCLUDE_MODE, the message does not contain (S, G) items; the record type is ALLOW_NEW_SOURCES, the current filter mode is INCLUDE, then the message contains (S, G) items; The type is BLOCK_OLD_SOURCES, and the current filtering mode is EXCLUDE, then the message contains (S, G) items.
{Leave, (S, G) }的判断准则可如下: 记录类型为 MODE— IS— INCLUDE 或者 CHANGE— TO— INCLUDE— MODE, 则报文不包含(S, G)项; 记录类型为 MODE— IS— EXCLUDE或者 CHANGE— TO— EXCLUDE— MODE , 则报文中不包含 ( S, G)项;记录类型为 ALLOW— NEW_SOURCES,当前过滤模式是 EXCLUDE, 则报文包含(S, G)项; 记录类型为 BLOCK_OLD— SOURCES, 并且当前过滤 模式是 INCLUDE, 则报文包含(S, G)项。 增加端口可包括以下步骤: 当端口的通道请求信息为 {Join, ( S , G ) }时, 搜索 (S, G)通道, 假如不匹配, 该 {Join, (S, G) }将被扔弃; 假如匹配, 在 (S, G)通道中增加端口。 删除端口可包括以下步骤: 当端口的通道请求信息为 {Leave, (S, G) } 时, 在(S, G)通道中删除端口。 重组 IGMP协议报文并转发给三层交换机可包括以下步骤: 二层交换机将 IGMP协议报文封装成多条 MODE— IS— INCLUDE记录 [G, SI], [G, S2], [G, S3] ,.·..· , 然后通过不同的通道, 将这些记录转发给三层交换机。 下面将参考附图 2并结合实施例, 来详细说明本发明, 如图 2所示, 包括 以下步骤: 步驟 S202. 在二层设备中划分一些 VLAN ( Virtual Local Area Network )资 源给组播专用, 并且划分该组播 VLAN的工作域。 步驟 S204. 在组播 VLAN的工作域中, 增加若干 (S, G)通道。 步骤 S206. 在组播 VLAN中对端口发送的 IGMP协议报文和(S, G)通 道进行匹配。 步骤 S208, 假如一致, 那么将端口加入该 (S, G)通道。 步骤 S210, 假如冲突, 则在(S, G)通道中删除该端口。 步骤 S212. 在组播 VLAN中, 根据与 (S, G)通道匹配的情况, 重组报 文给三层设备。 The judgment criterion of {Leave, (S, G) } can be as follows: If the record type is MODE-IS-INCLUDE or CHANGE_TO-INCLUDE-MODE, the message does not contain (S, G) items; the record type is MODE-IS — EXCLUDE or CHANGE— TO— EXCLUDE— MODE , the message does not contain ( S, G) items; the record type is ALLOW_NEW_SOURCES, the current filter mode is EXCLUDE, Then the message contains (S, G) items; the record type is BLOCK_OLD_SOURCES, and the current filtering mode is INCLUDE, then the message contains (S, G) items. Adding a port may include the following steps: When the channel request information of the port is {Join, (S, G) }, the (S, G) channel is searched, and if it does not match, the {Join, (S, G) } will be thrown. Discard; if matching, add a port in the (S, G) channel. Deleting a port can include the following steps: When the channel request information of the port is {Leave, (S, G) }, the port is deleted in the (S, G) channel. Reassembling IGMP protocol packets and forwarding them to Layer 3 switches can include the following steps: Layer 2 switches encapsulate IGMP protocol packets into multiple MODE-IS-INCLUDE records [G, SI], [G, S2], [G, S3 ] , . . . . . , then forward these records to the Layer 3 switch through different channels. The present invention will be described in detail below with reference to FIG. 2 and in conjunction with the embodiments. As shown in FIG. 2, the method includes the following steps: Step S202. Divide some VLAN (Virtual Local Area Network) resources into multicast for the Layer 2 device. And divide the working domain of the multicast VLAN. Step S204. In the working domain of the multicast VLAN, add several (S, G) channels. Step S206. Match the IGMP protocol packet sent by the port with the (S, G) channel in the multicast VLAN. In step S208, if they are consistent, the port is added to the (S, G) channel. In step S210, if there is a conflict, the port is deleted in the (S, G) channel. Step S212. In the multicast VLAN, the packet is reassembled to the Layer 3 device according to the matching with the (S, G) channel.
一般来说二层设备在数据交换中通过 VLAN ID ( VID) 和目的 MAC ( Destination Media Access Control )地址(DA)来识別数据流和进行数据交换。 在这种交换体系中, 本质上无法区分数据流的源地址, 不能直接提供 SSM服务。 本发明通过在二层交换机上分出两部分 VLAN资源, 其中一部分给组播数 据流使用称之为组播 VLAN, 与二层交换机上联的三层交换设备通过这些组播 VLAN把组播流送入二层交换机的上联口; 另一部分 VLAN给( S, G)通道使 用, 每一个(S, G)通道对应一个 SGVLAN, 通过网管配置将 SGVALN分配 给特定源地址和組地址的 (S, G)通道。 本发明通过对(S, G)通道的配置, 将(S, G)通道的組播流与二层转发流(VID, DA) 对应。 在组网应用时,每一个组播 VLAN对应上联三层设备一个相同的 VLAN接 口, 在二层交换机上的 (S, G)通道都隶属于一个组播 VLAN, 对于同一组地 址的多个通道, 如(Sl, G) , (S2, G) , (S3, G) , 这些相同组地址 G, 不同源地址 Sl, S2, S3通道配置在不同的組播 VLAN中, 这样(SI, G) , ( S2, G ) , (S3, G)分别属于组播 VLAN1, 組播 VLAN2, 组播 VLAN3。 在 IGMPv3协议报文处理过程中, 包括 IGMPv3报告报文和查询报文处理。 与二层交换机相连的终端主机将发送报告报文来表明自己想加入或者离开的(S, G)通道。一条完整的组记录主要包含三个部分,組记录类型,一个组地址(G), 一列源地址 (SI, S2, S3 ) IGMPv3组记录的类型共有如下 6种: Generally, a Layer 2 device identifies a data stream and exchanges data through a VLAN ID (VID) and a destination MAC (Destination Media Access Control) address (DA) in data exchange. In this switching system, the source address of the data stream cannot be distinguished in essence, and the SSM service cannot be directly provided. The present invention divides two parts of VLAN resources on the Layer 2 switch, and a part of the multicast data stream uses a multicast VLAN, and the Layer 3 switching device connected to the Layer 2 switch passes the multicast stream through the multicast VLAN. It is sent to the uplink port of the Layer 2 switch; another part of the VLAN is given to the (S, G) channel. Each (S, G) channel corresponds to an SGVLAN, and the SGVALN is assigned to the (S, G) channel of a specific source address and group address through the network management configuration. The present invention corresponds to the (S, G) channel multicast stream and the layer 2 forwarding stream (VID, DA) by configuring the (S, G) channel. In the networking application, each multicast VLAN corresponds to the same VLAN interface of the Layer 3 device. The (S, G) channel on the Layer 2 switch belongs to a multicast VLAN. Channels, such as (Sl, G), (S2, G), (S3, G), these same group address G, different source addresses Sl, S2, S3 channels are configured in different multicast VLANs, such (SI, G ), (S2, G), (S3, G) belong to multicast VLAN 1, multicast VLAN 2, and multicast VLAN 3. IGMPv3 report packets and query packets are processed during IGMPv3 packet processing. The terminal host connected to the Layer 2 switch will send a report message to indicate the (S, G) channel that you want to join or leave. A complete group record mainly consists of three parts, group record type, one group address (G), and one column source address (SI, S2, S3). The types of IGMPv3 group records are as follows:
1. 当前状态是包含模式( MODE— IS JNCLUDE ) 1. The current state is the inclusion mode (MODE_IS JNCLUDE)
2. 当前状态是排除模式( MODE— IS— EXCLUDE ) 2. The current state is the exclusion mode (MODE_IS—EXCLUDE)
3. 过滤模式切换到包含模式 ( CHANGE— TO— INCLUDE— MODE ) 3. Filter mode to include mode (CHANGE_TO_INCLUDE_MODE)
4. 过滤模式切换到排除模式( CHANGE一 TO— EXCLUDE— MODE ) 4. Switch the filter mode to the exclusion mode (CHANGE TO - EXCLUDE - MODE)
5. 源地址列表变换, 允许新的源地址( ALLOW— NEW— SOURCES ) 5. Source address list transformation, allowing new source addresses ( ALLOW - NEW - SOURCES )
6. 源地址列表变换, 屏蔽旧的源地址(BLOCK— OLD— SOURCES) 在不同的记录模式下, 組地址和源地址列表对应的 (S, G)通道针对端口 的权限是不同的, 不管是什么组合, 对于二层交换机上的 (S, G)通道而言, 只会产生两种结果, 其一是该端口允许接收(S, G)通道中的流量, 为该 (S, G)通道的通流端口; 其二是该端口禁止接收(S, G)通道中的流量, 为该(S, G)通道的禁流端口。 所以二层交换机在 IGMPv3报告报文处理过程中, 处理的核心问题就是解 析报文中所有的 (S, G)通道, 然后和二层交换机上配置的 (S, G)通道进行 匹配, 看该端口是通流端口还是禁流端口, 假如是通流端口则需要在( S , G ) 通道增加该端口, 否则删除该端口。 二层交换机在根据报告报文处理完自己的通道后, 将对报文进行重组, 将 报文记录 [G, SI, S2, S3...... ], 封装成若干记录 [G, SI], [G, S2], [G, S3] 等, 然后根据二层交换机中 (S, G)通道在组播 VLAN中的隶属情况, 通过不 同的组播 VLAN, 将这些记录转发给三层设备。 具体地说, 当二层交换机从主机收到组播地址为 G,源地址列表依次为 Sl, S2, S3 的 MODE— IS— INCLUDE记录时, 例如二层交换机从主机收到一 IGMP 报告报文记录, 记录类型为 MODE—ISJNCLUDE, 组播地址为 G, 源地址列表 依次为 SI, S2, S3, 假如该端口为 (SI, G) , (S2, G), (S3, G) 的通流 端口, 将把该记录分解成三条 MODE— IS— INCLUDE, (SI, G)记录将发送到 组播 VLANl接口, ( S2, G )记录将发送到组播 VLAN2接口, ( S3 , G )记录 将发送到组播 VLAN3接口。 这样(SI, G) , (S2, G) , (S3, G)的组播流 将通过不同的 VLAN1, 2, 3 发送到二层交换机, 对于相同组地址的数据流, VLANl , VLAN2, VLAN3分别对应不同的 VLAN ID,这样通过不同的 VLAN ID 来区分不同的源地址, 实现(S, G)和(VID, G) 的——对应关系。 实现二层 交换机的 SSM控制。 实现时, 首先在组播工作域中规定源端口和接收端口, 其中源端口为组播 VLAN中的节目来源端口,对于二层交换机而言,通常是和三层路由器相联的端 口; 接收端口是节目观看端口, 通常为节目流出端口, 这样非源端口和接收端口 的报文将被忽略。 然后, 以组记录为单位分解报告报文, 得到端口对若干通道的请求信息, 每一个通道构成为请求类型 (R: Request)加上通道信息(S, G) , 简写为 {R, (S, G) }, 请求类型有加入(Join)和离开(Leave)两种, 分别标为 {Join, (S, G) W{Leave, (S, G) }。 报文中 {Join, (S, G) }的判断准则如下: 6. Source address list conversion, masking the old source address (BLOCK_OLD_SOURCES) In different recording modes, the group address and the source address list correspond to the (S, G) channel's permissions for the port are different, whether it is What combination, for the (S, G) channel on the Layer 2 switch, only two results are produced. One is that the port allows the traffic in the (S, G) channel to be received, for the (S, G) channel. The traffic port; the second is that the port is prohibited from receiving traffic in the (S, G) channel, which is the forbidden port of the (S, G) channel. Therefore, the core problem of the Layer 2 switch in the processing of IGMPv3 report packets is to resolve all (S, G) channels in the packets and then match the (S, G) channels configured on the Layer 2 switch. Whether the port is a flow port or a no-flow port, if it is a flow port, it needs to be at (S, G) The port is added to the channel, otherwise the port is deleted. After processing the channel according to the report message, the Layer 2 switch reassembles the message and encapsulates the message record [G, SI, S2, S3...] into several records [G, SI ], [G, S2], [G, S3], etc., then forward these records to the third layer through different multicast VLANs according to the membership of the (S, G) channel in the Layer 2 switch in the multicast VLAN. device. Specifically, when the Layer 2 switch receives the multicast address as G from the host, and the source address list is MODE_IS-INCLUDE of S1, S2, and S3, for example, the Layer 2 switch receives an IGMP report message from the host. Record, record type is MODE-ISJNCLUDE, multicast address is G, source address list is SI, S2, S3, if the port is (SI, G), (S2, G), (S3, G) The port will be decomposed into three MODE-IS-INCLUDE, (SI, G) records will be sent to the multicast VLAN1 interface, (S2, G) records will be sent to the multicast VLAN2 interface, (S3, G) records will be Send to the multicast VLAN3 interface. Such (SI, G), (S2, G), (S3, G) multicast streams will be sent to the Layer 2 switch through different VLANs 1, 2, 3, for the same group address data stream, VLAN1, VLAN2, VLAN3 Corresponding to different VLAN IDs, so that different source IDs can be distinguished by different VLAN IDs, and the correspondence between (S, G) and (VID, G) is realized. Implement SSM control of Layer 2 switches. In the implementation, the source port and the receiving port are first specified in the multicast working domain, where the source port is a program source port in the multicast VLAN, and for the layer 2 switch, it is usually a port associated with the layer 3 router; It is the program viewing port, usually the program out port, so the packets of the non-source port and the receiving port will be ignored. Then, the report message is decomposed in units of group records, and the request information of the port for several channels is obtained. Each channel is configured as a request type (R: Request) plus channel information (S, G), abbreviated as {R, (S , G) }, the request type has two types: Join and Leave, which are labeled {Join, (S, G) W{Leave, (S, G) }. The judgment criteria of {Join, (S, G) } in the message are as follows:
1. 记 录 类 型 为 MODE— IS— INCLUDE 或 者 CHANGE— TO— INCLUDE— MODE, 报文包含(S, G)项; 1. The record type is MODE—IS—INCLUDE or CHANGE—TO—INCLUDE—MODE, and the message contains (S, G) items;
2. 记 录 类 型 为 MODE_IS— EXCLUDE 或 者 CHANGEJTO— EXCLUDE— MODE, 报文中不包含( S, G )项; 3. 记录类型为 ALLOW— NEW— SOURCES , 当前过滤模式是 INCLUDE , 报文包含(S, G)项; _; 2. The record type is MODE_IS—EXCLUDE or CHANGEJTO—EXCLUDE—MODE, and the message does not contain (S, G) items; 3. The record type is ALLOW-NEW-SOURCES, the current filter mode is INCLUDE, and the message contains (S, G) items; _ ;
4. 记录类型为 BLOCK— OLD— SOURCES , 并且当前过滤模式是 EXCLUDE, 报文包含(S, G)项; 报文中 {Leave, (S, G) }的判断准则如下: 4. The record type is BLOCK_OLD_SOURCES, and the current filter mode is EXCLUDE, the message contains (S, G) items; the judgment criteria of {Leave, (S, G) } in the message are as follows:
1. 记 录 类 型 为 MODE— ISJNCLUDE 或 者 CHANGE— TOJNCLUDE— MODE, 报文不包含(S, G) 项; 1. The record type is MODE—ISJNCLUDE or CHANGE—TOJNCLUDE—MODE, and the message does not contain (S, G) items;
2. 记 录 类 型 为 MODE— IS— EXCLUDE 或 者 CHANGE— TO— EXCLUDE— MODE, 报文中不包含(S, G)项; 2. The type of record is MODE—IS—EXCLUDE or CHANGE—TO—EXCLUDE—MODE, and the message does not contain (S, G) items;
3. 记录类型为 ALLOW一 NEW— SOURCES , 当前过滤模式是 EXCLUDE, 报文包含(S, G) 项; 3. The record type is ALLOW-NEW_SOURCES, the current filter mode is EXCLUDE, and the message contains (S, G) entries;
4. 记录类型为 BLOCK— OLD— SOURCES , 并且当前过滤模式是 INCLUDE, 报文包含(S, G)项; 具体流程如下: 步骤 1. 将接收端口接收到的 IGMPv3报告以组记录为单位进行分解, 得 到一系列 {Join, (S, G) p{Leave, (S, G) }, 然后逐个处理它们。 步骤 2. 当处理 {Join, (S, G) }时, 搜索二层交换机上配置的各个组播 VLAN 中配置的 (S, G)通道, 假如不匹配, 该 {Join, (S, G) }将被扔弃; 假如匹配, 在二层 (S, G)转发通道中增加该端口; 然后在 (S, G)从属的組 播 VLAN中, 重組记录, 发送到源端口; 步骤 3. 当处理 {Leave, ( S, G ) }时, 在二层( S, G )转发通道中删除该 端口; 然后在 (S, G)从属的组播 VLAN中, 修改报告报文, 组建删除该 (S, G)条目报文, 发送给源端口。 在 IGMPv3协议查询报文处理过程中, 二层交换机处理所有源端口过来的 IGMPv3查询报文。 收到的查询报文都将在组播 VLAN内发送到所有接收端口。 由以上描述可以看出, 本发明实现了以下有益效果, 本文提供一种二层交换机 IGMPv3 的实现方法, 是一种 "受控监听 ( Dominated IGMPv3 Snooping ) "方法,。该实现方法在任意源组播 IGMP Snooping 的基础上, 根椐二三层设备組播部署规划, 对 IGMP v3报文进行有效的分析处 理, 进行拆分, 转换, 记录, 再传递给上层路由器。 该方法不再是单纯的监听和 记录, 它需要根据组播流量的部署情况对 IGMPv3报文进行监听和处理。 该方法 有效地实现了二层交换机对 IGMPv3的支持,使二层交换机在现有奈件就可以具 备较强的特定源組播功能。 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。 凡在本发明的精神和原则之 内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明的保护范围之内。 4. The record type is BLOCK_OLD_SOURCES, and the current filter mode is INCLUDE, and the message contains (S, G) items. The specific process is as follows: Step 1. Decompose the IGMPv3 report received by the receiving port in units of group records. , get a series of {Join, (S, G) p {Leave, (S, G) }, and then process them one by one. Step 2. When {Join, (S, G) } is processed, search for the (S, G) channel configured in each multicast VLAN configured on the Layer 2 switch. If there is no match, the {Join, (S, G) } will be discarded; if it matches, add the port in the Layer 2 (S, G) forwarding channel; then in the (S, G) dependent multicast VLAN, reorganize the record and send it to the source port; Step 3. When When {Leave, ( S, G ) } is processed, the port is deleted in the Layer 2 (S, G) forwarding channel; then, in the (S, G) subordinate multicast VLAN, the report message is modified, and the deletion is established. S, G) The entry message is sent to the source port. During the processing of IGMPv3 Query messages, the Layer 2 switch processes IGMPv3 Query messages from all source ports. The received query messages will be sent to all receiving ports in the multicast VLAN. As can be seen from the above description, the present invention achieves the following beneficial effects. This document provides a method for implementing a Layer 2 switch IGMPv3, which is a "Dominated IGMPv3 Snooping" method. On the basis of any source multicast IGMP Snooping, the implementation method effectively analyzes and processes IGMP v3 packets based on the multicast deployment plan of the Layer 2 and Layer 3 devices, splits, converts, records, and delivers them to the upper-layer routers. This method is no longer a simple monitoring and recording. It needs to monitor and process IGMPv3 packets according to the deployment of multicast traffic. The method effectively implements the support of the Layer 2 switch for IGMPv3, so that the Layer 2 switch can have a strong specific source multicast function in the existing device. 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 spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims

权利要求书 Claim
1. 一种用于包括二层交换机和三层交换机的网络的特定源组播方法, 其特征在于, 包括以下步骤: A specific source multicast method for a network including a Layer 2 switch and a Layer 3 switch, comprising the steps of:
在所述二层交换机中划分 VLAN资源给特定源组播专用, 并 且划分所述 VLAN资源的工作域;  Allocating VLAN resources to the specific source multicast in the Layer 2 switch, and dividing the working domain of the VLAN resource;
在所述工作域中增加 (S, G )通道;  Adding (S, G) channels in the working domain;
在所述 VLAN资源中对端口发送的 IGMP协议报文和( S, G ) 通道进亍匹配; 以及  Matching the IGMP protocol packet sent by the port with the (S, G) channel in the VLAN resource;
根据匹配的情况, 重组所述报文并转发给所述三层交换机。  The packet is reassembled and forwarded to the Layer 3 switch according to the matching situation.
2. 根据权利要求 1所述的特定源组播方法, 其特征在于, 在所述工作 域中增加 (S, G )通道包括以下步骤: 2. The specific source multicast method according to claim 1, wherein the adding (S, G) channel in the working domain comprises the following steps:
在所述工作域中规定源端口和接收端口, 其中所述源端口为 所述 VLAN资源中的节目来源端口 , 所述接收端口是所述 VLAN 资源中的节目观看端口。  A source port and a receiving port are defined in the working domain, where the source port is a program source port in the VLAN resource, and the receiving port is a program viewing port in the VLAN resource.
3. 根据权利要求 2所述的特定.源组播方法, 其特征在于, 匹配包括以 下步骤: 3. The specific source multicast method according to claim 2, wherein the matching comprises the following steps:
假如所述 4艮文和( S , G )通道一致, 那么将所述端口加入( S , G )通道;  If the 4艮 and (S, G) channels are identical, then the port is added to the (S, G) channel;
如所述 4艮文和(S, G )通道冲突, 则在(S, G )通道中删 除所述端口。  If the 4艮 and (S, G) channels collide, the port is deleted in the (S, G) channel.
4. 根据权利要求 3所述的特定源组播方法, 其特征在于, 判断所述报 文和(S, G )通道是否一致包括以下步骤: The specific source multicast method according to claim 3, wherein determining whether the message and the (S, G) channel are consistent comprises the following steps:
解析所述报文中所有的通道请求信息, 然后和所述二层交换 机上配置的( S , G )通道进行匹配, 以此判断所述端口是通流端口 还是禁流端口, 假如是通流端口则是一致, 否则为沖突。 根据权利要求 4所述的特定源組播方法, 其特征在于, 解析所述报 文中所有的通道请求信息包括以下步骤: Parsing all the channel request information in the packet, and then matching with the (S, G) channel configured on the layer 2 switch, so as to determine whether the port is a traffic port or a no-flow port, if it is a traffic flow The ports are the same, otherwise they are conflicts. The specific source multicast method according to claim 4, wherein the parsing all channel request information in the packet comprises the following steps:
以组记录为单位分解所述报文得到所述通道请求信息, 其中, 将请求的每一个通道构成为请求类型 {R, (S, G) }, 将请求类型 分别标为 {Join, (S, G) {Leave, (S, G) }。 根据权利要求 5所述的特定源组播方法, 其特征在于, {Join, (S, G) }的判断准则如下: .  Decomposing the message in units of group records to obtain the channel request information, wherein each channel of the request is configured as a request type {R, (S, G) }, and the request type is respectively marked as {Join, (S , G) {Leave, (S, G) }. The specific source multicast method according to claim 5, wherein the criterion of {Join, (S, G) } is as follows:
i己 录 类 型 为 MODE—IS— INCLUDE 或 者 CHANGE— TO— INCLUDE— MODE , 则所述报文包含( S , G )项; i己 录 类 型 为 MODE—IS_EXCLUDE 或 者 CHANGE— TO— EXCLUDE— MODE , 则所述报文中不包含( S, G) 项;  If the record type is MODE-IS-INCLUDE or CHANGE_TO-INCLUDE-MODE, then the message contains (S, G); if the record type is MODE-IS_EXCLUDE or CHANGE_TO-EXCLUDE-MODE, then The message does not contain (S, G) items;
"¾录类型为 ALLOW— NEW— SOURCES , 当前过滤模式是 INCLUDE, 则所述报文包含(S, G)项;  "3⁄4 recording type is ALLOW-NEW_SOURCES, the current filtering mode is INCLUDE, then the message contains (S, G) items;
记录类型为 BLOCK—OLD_SOURCES, 并且当前过滤模式是 EXCLUDE, 则所述报文包含(S, G) 项。 根据权利要求 5所述的特定源组播方法, 其特征在于, {Leave, ( S, G) }的判断准则: ^下:  The record type is BLOCK_OLD_SOURCES, and the current filter mode is EXCLUDE, then the message contains (S, G) entries. The specific source multicast method according to claim 5, wherein the criterion of {Leave, (S, G) } is: ^下下:
i已 录 类 型 为 MODE— IS— INCLUDE 或 者 CHANGE— TO— INCLUDE— MODE, 则所述 4艮文不包含(S, G)项; 己 录 类 型 为 MODE— IS— EXCLUDE 或 者 CHANGE— TO— EXCLUDE— MODE , 则所述^ =艮文中不包含( S , G) 项;  If the recorded type is MODE—IS—INCLUDE or CHANGE—TO—INCLUDE—MODE, then the 4艮 text does not contain (S, G) items; the recorded type is MODE—IS—EXCLUDE or CHANGE—TO—EXCLUDE— MODE , then the ^ = 艮 text does not contain ( S , G) items;
"^已录类型为 ALLOW— NEW— SOURCES , 当前过滤模式是 EXCLUDE, 则所述报文包含(S, G) 项;  "^ has been recorded as ALLOW-NEW_SOURCES, and the current filtering mode is EXCLUDE, then the message contains (S, G) items;
记录类型为 BLOCK— OLD_SOURCES, 并且当前过滤模式是 INCLUDE, 则所述报文包含(S, G)项。 The record type is BLOCK_OLD_SOURCES, and the current filter mode is INCLUDE, then the message contains (S, G) entries.
8. 根据权利要求 5所述的特定源組播方法, 其特征在于, 增加所述端 口包 4舌以下步骤: 8. The specific source multicast method according to claim 5, wherein the step of adding the port package is as follows:
当所述端口的所述通道请求信息为 {Join, (S, G) }时, 搜索 所述(S, G)通道, ·ί叚如不匹配, 所述 {Join, (S, G) }将被扔弃; £如匹配, 在所述(S, G)通道中增加所述端口。  Searching for the (S, G) channel when the channel request information of the port is {Join, (S, G) }, if the channel does not match, the {Join, (S, G) } Will be discarded; £ as matched, add the port in the (S, G) channel.
9. 根据权利要求 5所述的特定源組播方法, 其特征在于, 删除所述端 口包括以下步骤: 9. The specific source multicast method according to claim 5, wherein deleting the port comprises the following steps:
当所述端口的所述通道请求信息为 {Leave, (S, G) }时, 在 所述(S, G)通道中删除所述端口。  When the channel request information of the port is {Leave, (S, G) }, the port is deleted in the (S, G) channel.
10. 根据权利要求 5 所述的特定源组播方法, 其特征在于, 重组所述 IGMP协议^ 1艮文并转发给所述三层交换机包括以下步驟: 10. The Source-Specific Multicast method according to claim 5, wherein the recombinant ^ 1 Gen IGMP protocol and forwards to the three switches comprising the steps of:
所述二层交换机将所述 IGMP 协议报文封装成多条 MODE— IS— INCLUDE记录 [G, SI], [G, S2], [G, S3]……, 然后 通过不同的所述通道, 将这些记录转发给所述三层交换机。  The Layer 2 switch encapsulates the IGMP protocol message into multiple MODE-IS-INCLUDE records [G, SI], [G, S2], [G, S3], ... and then through different channels. These records are forwarded to the Layer 3 switch.
PCT/CN2006/003101 2006-11-17 2006-11-17 A specific source multicasting method WO2008058428A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN200680055256.6A CN101480010B (en) 2006-11-17 2006-11-17 A specific source multicasting method
PCT/CN2006/003101 WO2008058428A1 (en) 2006-11-17 2006-11-17 A specific source multicasting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2006/003101 WO2008058428A1 (en) 2006-11-17 2006-11-17 A specific source multicasting method

Publications (1)

Publication Number Publication Date
WO2008058428A1 true WO2008058428A1 (en) 2008-05-22

Family

ID=39401304

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2006/003101 WO2008058428A1 (en) 2006-11-17 2006-11-17 A specific source multicasting method

Country Status (2)

Country Link
CN (1) CN101480010B (en)
WO (1) WO2008058428A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102185785A (en) * 2011-06-07 2011-09-14 杭州华三通信技术有限公司 Layer 2 forwarding chip and realization method thereof
CN106302188A (en) * 2015-05-18 2017-01-04 中兴通讯股份有限公司 The multicast message transmission control method of a kind of switch device and device

Citations (4)

* 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
CN1555165A (en) * 2003-12-26 2004-12-15 IP multicasting precision port repeat method in three layer exchanging
CN1852232A (en) * 2005-11-29 2006-10-25 华为技术有限公司 Multi-cast method of specific source and irrespective source multi-cast cooperation and its access-in apparatus
CN1863153A (en) * 2005-10-09 2006-11-15 华为技术有限公司 Method of ethernet supporting source specific multicast forwarding and apparatus thereof

Patent Citations (4)

* 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
CN1555165A (en) * 2003-12-26 2004-12-15 IP multicasting precision port repeat method in three layer exchanging
CN1863153A (en) * 2005-10-09 2006-11-15 华为技术有限公司 Method of ethernet supporting source specific multicast forwarding and apparatus thereof
CN1852232A (en) * 2005-11-29 2006-10-25 华为技术有限公司 Multi-cast method of specific source and irrespective source multi-cast cooperation and its access-in apparatus

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102185785A (en) * 2011-06-07 2011-09-14 杭州华三通信技术有限公司 Layer 2 forwarding chip and realization method thereof
CN106302188A (en) * 2015-05-18 2017-01-04 中兴通讯股份有限公司 The multicast message transmission control method of a kind of switch device and device

Also Published As

Publication number Publication date
CN101480010A (en) 2009-07-08
CN101480010B (en) 2012-06-27

Similar Documents

Publication Publication Date Title
US7835276B2 (en) Admission control mechanism for multicast receivers
US8339973B1 (en) Multicast traceroute over MPLS/BGP IP multicast VPN
US7590120B2 (en) Enhanced multicast VLAN registration
US8422499B2 (en) Methods and apparatus for managing multicast traffic
CN101616014B (en) Method for realizing cross-virtual private local area network multicast
CN100433730C (en) Method and system of multicast and video-on-demand
EP1318628A1 (en) Method and apparatus of directing multicast traffic in an Ethernet MAN
CN100442770C (en) Method for realizing muti-casting in BGP/MPLS VPN
WO2010075771A1 (en) Extranet networking method, system and device for multicast virtual private network
WO2006063513A1 (en) A method for realizing multicasting in a virtual private network using a virtual router
JP2003032287A (en) Method and apparatus for connecting networks, and system using the apparatus
WO2007056923A1 (en) Method for implementing multicast
CN102215172B (en) A kind of method and system for realizing cross-virtual private local area network multicast
US9264244B2 (en) Multicasting in a provider backbone bridge network
WO2007059679A1 (en) A method for processing the abnormal multicast service and a network equipment thereof
US8559353B2 (en) Multicast quality of service module and method
WO2011020346A1 (en) Method and apparatus for forwarding multicast data
WO2007054006A1 (en) A base station of wideband wireless access system and a method for realizing group-broadcast service
WO2009024054A1 (en) A method, device and system for realizing the management protocol agent for members in a multicast group
US20060029001A1 (en) Multicast source discovery
CN101083664A (en) Method for preventing dual-layer multicasting data stream flood
WO2007041942A1 (en) System for Ethernet supporting the transmitting of the source-specific multicast and the method thereof
WO2008058428A1 (en) A specific source multicasting method
WO2007059699A1 (en) Method and date relay entity for relaying the date frame by the date relay entity
WO2008141516A1 (en) Message transmitting method, transmitting device and transmitting system

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200680055256.6

Country of ref document: CN

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

Ref document number: 06805250

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06805250

Country of ref document: EP

Kind code of ref document: A1