WO2014117530A1 - 一种trill分发树故障的通告方法和装置 - Google Patents

一种trill分发树故障的通告方法和装置 Download PDF

Info

Publication number
WO2014117530A1
WO2014117530A1 PCT/CN2013/083916 CN2013083916W WO2014117530A1 WO 2014117530 A1 WO2014117530 A1 WO 2014117530A1 CN 2013083916 W CN2013083916 W CN 2013083916W WO 2014117530 A1 WO2014117530 A1 WO 2014117530A1
Authority
WO
WIPO (PCT)
Prior art keywords
distribution tree
failure
group
trill
trill distribution
Prior art date
Application number
PCT/CN2013/083916
Other languages
English (en)
French (fr)
Inventor
代雪会
廖婷
翟洪军
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2014117530A1 publication Critical patent/WO2014117530A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4604—LAN interconnection over a backbone network, e.g. Internet, Frame Relay
    • H04L12/462—LAN interconnection over a bridge based backbone
    • H04L12/4625—Single bridge functionality, e.g. connection of two networks over a single bridge
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/48—Routing tree calculation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/66—Layer 2 routing, e.g. in Ethernet based MAN's
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
    • H04L43/0811—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking connectivity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route

Definitions

  • the present invention relates to the field of network communication technologies, and in particular, to a method and apparatus for reporting TRILL distribution tree failures. Background technique
  • TRILL Transparent Interconnection over Lots of Links
  • L2MP Layer 2 Multiple Path
  • the prior art implements L2MP by introducing an intermediate system to intermediate system (IS-IS) routing protocol into a Layer 2 network.
  • IS-IS intermediate system to intermediate system
  • RB route bridge
  • an alias is uniquely identified by an alias (Nickname).
  • Ingress the routing bridge responsible for encapsulating the original data frame of the end device into the TRILL format and injecting into the TRILL network
  • Ingress the encapsulating the original data frame into the TRILL format refers to the original data frame.
  • the TRRILL header and the external frame header are added before, and the TRILL header mainly includes the Nickname and the hop count of the entry and exit RBs of the TRILL network; at the exit of the TRILL network, the TRILL data frame is decapsulated into the original data frame and forwarded to the end device.
  • the routing bridge is called an Egress.
  • the Egress learns which Ingress is imported into the TRILL of the data frame, and forms a Medium Access Control (MAC) information table ⁇ D_MAC, Ingress Nickname, ... ⁇ . Because the Ingress and Egress routing bridges are in the edge of the TRILL network, they are also called Edge Routing Bridges.
  • MAC Medium Access Control
  • the RB can serve one end system in any virtual local area network (VLAN).
  • the RB is called the service provider of this end system, such as The specified forwarder of the VLAN on the shared link (AF, Appointed Forwarder).
  • AF Appointed Forwarder
  • This can effectively avoid loops, but it also brings some problems.
  • the Ingress- Nickname changes in some MAC entries on the remote Egress; the end-system links multi-homed through the point-to-point link ( Multi-homing)
  • Multi-homing When multiple RBs are connected, the load balancing on the member link causes frequent fluctuations in the Ingress- Nickname of some MAC entries on the remote RB (called flip-flop), causing out-of-order traffic.
  • the TRILL working group proposed the concept of a Routing Bridge Group (RBG) or a Virtual Route Bridge (RBv).
  • RBG Routing Bridge Group
  • RBv Virtual Route Bridge
  • the group members share a Nickname, called the group Nickname.
  • the group Nickname When the RBv member imports the data frame into the TRILL network, the data frame is encapsulated by the group Nickname, thereby solving the above flip-flop problem.
  • TRILL defines a distribution tree for forwarding multicast, broadcast, and unknown unicast packets.
  • One or more distribution trees can be established in the TRILL campus. These distribution trees are all RB devices that cover the TRILL network. By selecting a root, the distribution of all RBs in the TRILL network is calculated. Tree, the distribution tree is VLAN shared. In addition, you can perform different pruning operations on the distribution tree for different VLANs to avoid sending packets to unnecessary RBs. When an RBv exists, all the member RBs receive the multicast stream on the distribution tree because the distribution tree of the TRILL network covers all member devices of the RBv.
  • the Coordinated Multicast Trees (CMT) and the pseudo nodes (PN, Pseudonode) drafts stipulate that only one of the RBs will be distributed by one of the members.
  • the multicast stream received on the tree is forwarded to the end system, which may be referred to as the designated forwarder of the distribution tree.
  • the PN draft also provides a protection mechanism for downlink multicast forwarding in the case of member link failure of the RBv-to-end system, that is, TRILL in a LAG.
  • All the distribution trees calculated by the network allocate the downlink designated forwarders among the members of the LAG. If the member RB detects that a fault has occurred on the corresponding LAG member link, the member RB sends an announcement message to notify other member RBs of the RB. Place All the multicast trees in the sub-area re-elect the new designated forwarders in the downstream direction.
  • this method only covers the scenario where the LAG access side is faulty, and the member RB in the RBv detects how the fault occurs on the distribution tree on the TRILL network side, and the prior art cannot provide the problem. s solution.
  • a RBv on a LAN allocates a unique AF on the LAN according to the VLAN, and when an AF does not receive the desired multicast connection along a distribution tree, After the check packet, the connectivity of the AF on the distribution tree is faulty, and the prior art cannot provide a solution to the fault. That is to say, the prior art cannot provide a solution for notifying the TRILL distribution tree fault information between member RBs in the RBv scenario.
  • the main purpose of the embodiments of the present invention is to provide a method and an apparatus for notifying a TRILL distribution tree fault, so as to advertise a TRILL distribution tree fault information between member RBs in an RBv scenario.
  • the embodiment of the invention provides a method for notifying a fault of a multi-link transparent interconnection TRILL distribution tree, and the method includes:
  • the member routing bridge RB in the group detects a failure in the TRILL distribution tree, the member RB notifies the other member RBs in the group in which the member RB is located.
  • the method also includes:
  • the other member RBs that receive the fault information in the group reselect the designated forwarder AF for the TRILL distribution tree according to the received fault information.
  • Distributing 4 pairs of reselected AFs for the TRILL by: randomly selecting from among the available member RBs in the group or available member RBs in the group to negotiate from among the available member RBs in the group Choose.
  • the member routing bridge RB in the group detects that a fault occurs on the TRILL distribution tree, and includes: one or more ingress RBs in the TRILL distribution tree respectively send multicast connectivity detection messages along the TRILL distribution tree.
  • the member RB in the group does not receive the multicast connectivity check packet of the corresponding ingress RB, it is determined that the packet forwarding of the corresponding ingress RB to the member RB is faulty.
  • the fault information includes: an RB identifier that detects a fault, an identifier of the group to which the RB that detected the fault belongs, a TRILL distribution tree identifier that is faulty, and an ingress RB that has a packet forwarding fault between the RBs that detected the fault.
  • the group is a virtual routing bridge RBv.
  • the identifier of the group to which the RB belongs is the alias Nickname of the RBv.
  • the embodiment of the invention further provides a notification device for the multi-link transparent interconnection TRILL distribution tree fault, which is applicable to the member routing bridge RB in the group, including:
  • the fault detection module is configured to detect a fault on the TRILL distribution tree.
  • the fault notification module is configured to notify the other member RBs in the group where the member RB is located when the fault occurs on the TRILL distribution tree.
  • the device also includes:
  • Specify the forwarder AF selection module configured to reselect AF according to the received fault information for the distribution tree described in TRILL.
  • the AF selection module is further configured to reselect the AF for the TRILL distribution tree by randomly selecting or negotiating among the available member RBs within the group from among the available member RBs within the group.
  • the fault detection module is further configured to determine, when the multicast connectivity detection packet sent by the corresponding ingress RB on the TRILL distribution tree is received, the packet of the corresponding ingress RB to the member RB. The forwarding has failed.
  • the fault information includes: an RB identifier that detects a fault, and an RB that detects the fault.
  • the identity of the group the TRILL distribution tree identifier of the failure, and the identity of the ingress RB with the packet forwarding failure between the RBs that detected the failure.
  • the bridge group virtual routing RBv, corresponding, RB identity belongs to another group of 1 J RBv the name of Nickname
  • the method and device for notifying the TRILL distribution tree fault in the RBv scenario can be used to notify the TRILL distribution tree fault information between the member RBs in the RBv scenario.
  • the fault information is notified to the member RBs in the group, and does not need to be flooded to all the RBs.
  • FIG. 1 is a schematic flow chart of a method for notifying a fault of a TRILL distribution tree according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a scenario in which an RB belongs to an RBv in a LAG scenario according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a scenario in which a RB belongs to two RBvs in a LAG scenario according to an embodiment of the present invention
  • FIG. 4 is a schematic diagram of a LAN scenario according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a device for notifying a TRILL distribution tree fault according to an embodiment of the present invention. detailed description
  • FIG. 1 A method for notifying a fault of a TRILL distribution tree according to the embodiment of the present invention is as shown in FIG. 1 , which mainly includes:
  • Step 101 When the member RB in the group detects that a fault occurs on the TRILL distribution tree, the member RB notifies the other member RBs in the group where the member RB is located.
  • the method may further include:
  • Step 102 The other member RBs that receive the fault information in the group reselect the AF for the TRILL distribution tree according to the received fault information. Preferably: selection from among the available member RBs within the group, from a random selection in a member RB available within the group or an available member RB negotiation within the group.
  • the method further includes: sending, by the ingress RB of the TRILL distribution tree, the multicast connectivity detection packet along the TRILL distribution tree, where the member RBs in the group cannot receive the multicast of the corresponding ingress RB When the connectivity check packet is received, it is determined that the packet forwarding of the corresponding ingress RB to the member RB fails.
  • the fault information includes: an RB identifier that detects a fault, an identifier of the group to which the RB that detected the fault belongs, a TRILL distribution tree identifier that is faulty, and an ingress RB that has a packet forwarding fault between the RBs that detected the fault.
  • the group is an RBv
  • the identifier of the group to which the RB belongs is the Nickname of the RBv
  • the identifier of the group to which the RB belongs is uniquely identifies the group.
  • RB4, RB6, RB11, and RB12 are ingress RBs in FIG. 2.
  • RB4, RB6, RBI 1, and RBI 2 respectively send multicast connectivity detection reports.
  • the detection of the connectivity of the distribution tree wherein the ingress nickname of the TRILL header of the multicast connectivity detection message sent by each ingress RB is the nickname of each ingress RB, and the egress nickname is the nickname of RBI 0;
  • the distribution tree shown by the thick solid line, RB3 is the downlink designated forwarder of the distribution tree, and RB1, RB2, and RB3 belong to one group RBv1.
  • the embodiment of the present invention is directed to the case where a member RB shown in FIG. 2 belongs to only one RBv, and the method for notifying the TRILL distribution tree failure mainly includes:
  • Step 1 RB4, RB6, RB1, and RB12 periodically send a multicast connectivity detection message to detect connectivity of the distribution tree.
  • Step 2 If the RB8 device fails, RB1 and RB2 cannot receive the multicast connectivity detection message sent by RB11 and RB12 within a certain period of time. RB3 cannot receive RB4, RB6, RB1, and RB12 in a certain period of time. Multicast connectivity detection message.
  • Step 3 RB1, RB2, and RB3 respectively send a failure notification message to other member RBs in RBvl.
  • the fault message sent by RB1 carries ⁇ RB1, RBvl, RB10, RB1 l, RB12 ⁇ , and the RBI carried in the fault message indicates that the RBI is a node that detects a fault of the distribution tree, and RBv1 indicates that RB1 belongs to a member RB of RBv1, and RB10 represents The root of the tree, RB11 and RB12, indicates that the forwarding of the RB11 and RB12 to the RB1 in the distribution tree where the RB10 is the root of the tree has failed; the above fault information can be carried by ESADI, and the ESADI frame can be separately sent to the unicast mode.
  • the fault message sent by RB2 carries ⁇ RB2, RBvl, RB10, RBl l, RB12 ⁇ .
  • the RB2 carried in the fault message indicates that RB2 is the node that detects the fault of the distribution tree, RBv1 indicates that RB2 belongs to the member RB of RBvl, and RB10 represents the root of the tree.
  • RB11 and RB12 the above fault information can also be carried by ESADI, and the ESADI frame can be separately sent to RB1 and RB3 by unicast.
  • the RB3 sends a fault message carrying ⁇ RB3, RBv1, RB10, RB4, RB6, RB1, RB12 ⁇ .
  • the RB3 carried in the fault message indicates that RB3 is a node that detects a fault of the distribution tree, and RBv1 indicates that RB3 belongs to a member RB of RBvl.
  • RB10 indicates the root of the tree
  • RB4, RB6, RB1, and RB12 indicate that RB4, RB6, RB1, and RB12 have failed to forward to the RB3 in the distribution tree where the RB 10 is the root of the tree; here, the above fault information can also be carried by ESADI.
  • ESADI frames can be sent to RB1 and RB2 separately by unicast. Re-select a designated forwarder in the distribution tree where the RB 10 is the root.
  • RB1 and RB2 are available to the RBs in the RBv1 group for RB4 and RB6, and no member RBs are available in the RBv1 group for RB11 and RB12. Then, in this embodiment, only one AF of RB4 and RB6 can be reselected in the member RBs available in the RBv1 group: RB1 and RB2.
  • the designated forwarder may be reselected in a random manner, or the designated forwarder may be reselected by using the member RB negotiation in the group, and a specific algorithm may be used to reselect the designated forwarder.
  • the second embodiment of the present invention is also directed to the case where a member RB shown in FIG. 2 belongs to only one RBv, and the method for notifying the TRILL distribution tree failure mainly includes:
  • Step 1 RB4, RB6, RB11, and RB12 periodically send a multicast connectivity detection message to detect connectivity of the distribution tree.
  • Step 2 If the link of RB10-RB8 fails, RB3 cannot receive the multicast connectivity detection message sent by RB4 and RB6 within a certain period of time, but can receive the RBI 1 and RB12 sent along the distribution tree.
  • the multicast connectivity detection packet is received by the RB1 and the RB2, and the multicast connectivity detection message sent by the RB4 and the RB6 is received by the RB1 and the RB2.
  • Step 3 RB1, RB2, and RB3 respectively send a failure notification message to other member RBs in RBvl.
  • the fault information sent by RB3 carries ⁇ RB3, RBvl, RBIO, RB4, RB6 ⁇ , RBIO indicates that the root tree is faulty on the distribution tree of RBIO, and RB4 and RB6 represent the faults caused by the ingress RB.
  • the packet cannot be forwarded to the specified receiver RB3.
  • the fault information is advertised to other members in the group in RBvl, namely RB1 and RB2, by unicast ESADI. After receiving the fault information, RBI and RB2 find that RB3 is RBvl.
  • the RB1 and RB2 need to re-select the unique designated forwarder in the RBv1 when the RB4 and RB6 are imported as the ingress.
  • RB3 is not the designated forwarder
  • RBI is the designated forwarder of the distribution tree in RBvl
  • RB3 detects that the branch to the local node in the distribution tree has failed, the advertised information is sent out, after RB1 and RB2 are received.
  • the RB3 is not the designated forwarder, the RB3 does not need to re-elect the election.
  • the multicast traffic of the distribution tree is forwarded along RB1, and the normal forwarding of the existing forwarder is not affected.
  • the designated forwarder may be reselected in a random manner, or the designated forwarder may be reselected by using the member RB negotiation in the group, and a specific algorithm may be used to reselect the designated forwarder.
  • the fault information sent by RB1 and RB2 carries ⁇ RB1, RBvl, RB10, RB11, RB12 ⁇ and ⁇ RB2, RBvl, RB10, RB11, RB12 ⁇ .
  • RB3 is elected as ingress for RB11 and RB12.
  • the designated forwarder when forwarding the multicast stream for the RB4 and RB6 distribution tree along the RB10 is one of RB1 and RB2; the distribution of RB11 and RB12 is rooted along RB10.
  • the specified forwarder when the tree forwards the multicast stream is RB3.
  • the member elects a new multicast stream for the ingress RB and the distribution tree when it determines that it can also receive the multicast stream forwarded by the ingressRB along the tree.
  • the designated forwarder in this case, can ensure that available members forward the received multicast stream to the connected multicast receiver.
  • the third embodiment of the present invention is directed to the case where one member RB shown in FIG. 3 belongs to two RBvs.
  • the multicast stream sent by CE4 shown in FIG. 3 enters the TRILL network as an ingress RB through RB4, and RB4 selects RB10 as the root of the tree.
  • a distribution tree forwards the multicast stream, as shown by the thick solid line shown in Figure 3.
  • RB2 belongs to two RBvs at the same time, RBI and RB2 form RBv2, RB2 and RB3 form RBvl, and RB2 is simultaneously elected or configured as a designated forwarder of the distribution tree in RBvl and RBv2.
  • the notification methods for TRILL distribution tree faults mainly include: Step 1: RB4 periodically sends a multicast connectivity detection message to detect connectivity of the distribution tree.
  • Step 2 If the link of RB7-RB2 fails, RB2 cannot receive the multicast connectivity detection message sent by RB4 within a certain period of time.
  • Step 3 RB2 sends a failure notification message to RBv1 and other member RBs in RBv2.
  • the message sent by RB2 is two, one is sent to RB 1, and the carried information is ⁇ RB2, RB10, RBv2 ⁇ .
  • RBI finds that it is the only available member RB in RBv2, then The self-election is the designated forwarder of the distribution tree in RBv2; the other message is sent to RB3, and the carried information is ⁇ RB2, RB10, RBvl ⁇ .
  • RB3 finds that it is the only available member in RBvl.
  • RB then elects itself as the designated forwarder of the distribution tree in RBvl.
  • the fourth embodiment of the present invention is directed to the LAN scenario shown in FIG. 4.
  • RB1, RB2, and RB3 are AFs of VLANs 1-100, 101-200, and 201-300 on the LAN, respectively.
  • the notification methods for TRILL distribution tree faults mainly include:
  • Step 1 RB4 periodically sends a multicast connectivity detection message to detect the connectivity of the distribution tree.
  • the multicast BFD packet carries the VLAN 100, indicating that the connectivity of the prune tree of the VLAN 100 is checked.
  • Step 2 If the link of RB7-RB1 fails, RB1 cannot receive the multicast connectivity detection message sent by RB4 within a certain period of time.
  • the multicast tree transmitted by the multicast BFD check packet of VLAN 1-99 is not the multicast tree as shown by the solid line in Figure 4.
  • RB7-RB1 fails, RB1 only has multicast BFD of VLAN 100. The message is not received.
  • Step 3 RB1 sends a fault notification message to other member RBs on the LAN.
  • the fault information advertisement here may be sent by the multicast ESADI on the LAN or the intermediate system-intermediate system session (IIH, IS-IS HELLO), and the information carried in the message is ⁇ VLAN100 ⁇ .
  • the other members on the LAN receive the fault information sent by RB1, Knowing that the RBI is the AF of the VLAN, then re-allocating a new AF to VLAN 100.
  • This example is only an example in the LAN scenario. If the multicast connectivity check packet does not carry the VLAN information or carries the default VLAN, the RB that receives the multicast connectivity check packet fails to be sent. The information does not need to carry VLAN information.
  • each member on the LAN knows the AF information of all VLANs, when other members receive the fault information, they will re-allocate all the VLANs with the RB as the AF.
  • the RBs are used as new AFs (ie, it is determined which VLANs are the AFs as the AFs, and new AFs are reassigned to the remaining member RBs for these VLANs).
  • the embodiment of the present invention further provides a TRILL distribution tree fault notification device, which is applicable to member RBs in the group, as shown in FIG. 5, the device is mainly The method includes: a fault detection module 10 and a fault notification module 20;
  • the fault detection module 10 is configured to detect a fault on the TRILL distribution tree.
  • the fault notification module 20 is configured to notify the fault information of the member RB when the fault occurs on the TRILL distribution tree. Other members RB.
  • the apparatus further comprises: an AF selection module 30 configured to reselect the AF for the distribution tree based on the received fault information.
  • an AF selection module 30 configured to reselect the AF for the distribution tree based on the received fault information.
  • the AF selection module 30 is further configured to reselect the AF for the TRILL distribution tree by randomly selecting or negotiating from among the available member RBs in the group from among the available member RBs in the group.
  • the fault detection module 10 is further configured to: when the multicast connectivity detection packet sent by the corresponding ingress RB on the TRILL distribution tree is not received, determine the corresponding ingress RB to the member RB. The packet forwarding failed.
  • the fault information includes: an RB identifier that detects a fault, an identifier of the group to which the RB that detected the fault belongs, a TRILL distribution tree identifier that is faulty, and a packet forwarding fault between the RB and the faulty RB.
  • the identity of the ingress RB is not limited to: an RB identifier that detects a fault, an identifier of the group to which the RB that detected the fault belongs, a TRILL distribution tree identifier that is faulty, and a packet forwarding fault between the RB and the faulty RB.
  • the group is an RBv, and correspondingly, the identifier of the group to which the RB belongs is the RBv.
  • the fault detection module and the AF selection module may be a central processing unit (CPU), a processor (MPU, a Micro Processing Unit), and a digital signal processor (DSP) in the member RB.
  • CPU central processing unit
  • MPU Micro Processing Unit
  • DSP digital signal processor
  • FPGA programmable logic array
  • the failure notification module may be implemented by hardware having a communication function in the member RB.

Landscapes

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

Abstract

本发明公开了一种多链接透明互连（TRILL）分发树故障的通告方法和装置，方法包括：当组内的成员路由网桥（RB）检测到TRILL分发树上发生故障时，所述成员RB将相应的故障信息通知给所述成员RB所在组内的其他成员RB。通过本发明，能够实现在RBv场景下成员RB之间通告TRILL分发树故障信息。

Description

一种 TRILL分发树故障的通告方法和装置 技术领域
本发明涉及网络通信技术领域， 特别是指一种 TRILL分发树故障的通 告方法和装置。 背景技术
多链接透明互连 ( TRILL, Transparent Interconnection over Lots of Links ) 用于解决数据中心大二层网络中的多路径问题， 或称 L2MP ( Layer 2 Multiple Path ) 问题。 现有技术通过将中间系统对中间系统 （ IS-IS， Intermediate System to Intermediate System )路由协议引入二层网络实现了 L2MP。在 TRILL网络中，运行 TRILL协议的设备称为路由网桥（ RB， Route Bridge ), 且由别名 （Nickname )唯一的标识一台 RB。 在 TRILL网络的入 口， 负责将端设备 ( End Station ) 的原始数据帧封装成 TRILL格式并注入 TRILL网络的路由网桥称为 Ingress, 所述将原始数据帧封装成 TRILL格式 是指在原始数据帧前添加 TRRILL 头和外部帧头， TRILL 头中主要包括 TRILL网络的入口和出口 RB的 Nickname和跳数； 在 TRILL网络的出口， 负责将 TRILL数据帧解封为原始数据帧并转发给端设备的路由网桥称为 Egress, 同时， Egress还会学习所述数据帧是哪个 Ingress导入 TRILL的， 并形成介质访问控制 （MAC， Medium Access Control )信息表 {D_MAC， Ingress Nickname, …… }。 由于 Ingress和 Egress路由网桥在 TRILL网络中 处在边缘位置， 因此又被称为边缘路由网桥（ Edge Route Bridge )。
为了避免环路，在 TRILL网络的边界，在任何一个虚拟局域网（ VLAN, Virtual Local Area Network ) 内只能由一个 RB为一个端系统提供服务， 该 RB称为此端系统的服务提供者，如共享链路上 VLAN的指定转发者（ AF， Appointed Forwarder )。这样虽然能有效的避免环路，但也会带来一些问题， 如： AF切换后， 远端 Egress上某些 MAC表项中的 Ingress— Nickname发生 变化； 在端系统通过点对点链路多归属 （ Multi-homing )到多个 RB时， 成 员链路上的负荷分担会引起远端 RB上某些 MAC表项的 Ingress— Nickname 频繁波动 （称为 flip-flop ), 从而引起回来的流量的乱序与丟包， 导致会话 中断。为此， TRILL工作组提出了路由网桥组（ RBG， Route Bridge Group ) 或称为虚拟路由网桥（RBv， Virtual Route Bridge )的概念。 在 RBv内， 组 员共享一个 Nickname,称为组 Nickname, RBv成员在将数据帧导入 TRILL 网络时， 用组 Nickname封装数据帧， 从而解决了上述 flip-flop问题。
另外， TRILL 定义了分发树来实现对组播、 广播以及未知单播报文的 转发。 TRILL campus ( TRILL网络） 内可以建立一个或多个分发树， 这些 分发树是覆盖到 TRILL网络的所有 RB设备的；通过选择一个树根 ( Root )， 来计算到达 TRILL网络内的所有 RB的分发树，该分发树是 VLAN共享的。 另外， 也可以针对不同 VLAN对分发树进行不同的剪枝操作， 以避免将报 文发送到不必要的 RB。 当有 RBv存在时， 由于 TRILL网络的分发树会覆 盖 RBv的所有成员设备， 因此， 所有的成员 RB都会收到分发树上的组播 流。
为了避免连接到 RBv的端系统收到多份组播流， 协调组播树 ( CMT, Coordinated Multicast Trees )和伪节点 （PN， Pseudonode )草案都规定了只 能由其中的一个成员 RB将从分发树上收到的组播流转发到端系统，该成员 RB可以被称为该分发树的指定转发者。
针对链路聚合组（LAG， Link Aggregation Group ) 的场景， PN草案还 提供了一种 RBv到端系统的成员链路故障情况下下行组播流转发的保护机 制， 即在一个 LAG内， 为 TRILL网络计算出来的全部分发树在 LAG的成 员间分配下行指定转发者， 如果成员 RB检测到对应的 LAG成员链路上发 生了故障， 则该成员 RB发出通告消息来通知其它的成员 RB针对该 RB所 分的全部组播树在下行方向再重新选举出新的指定转发者。 但这种方式仅 仅是覆盖了 LAG接入侧故障的场景， 而针对 RBv 内的成员 RB检测到 TRILL 网络侧的分发树上出现了故障的情况该如何处理， 现有技术还无法 提供针对该问题的解决方案。
另外， 针对局域网 （ LAN， Local Area Network )的场景， 一个 LAN上 的 RBv内根据 VLAN来分配相应的 LAN上唯一的 AF， 当某个 AF接收不 到期望的沿某分发树过来的组播连通性检查报文后，表明该 AF在该分发树 上的连通性出现了故障， 现有技术还无法提供针对该故障的解决方案。 也 就是说， 现有技术还无法提供一种 RBv场景下成员 RB之间通告 TRILL分 发树故障信息的解决方案。 发明内容
有鉴于此， 本发明实施例的主要目的在于提供一种 TRILL分发树故障 的通告方法和装置， 以实现在 RBv场景下成员 RB之间通告 TRILL分发树 故障信息。
为达到上述目的， 本发明实施例的技术方案是这样实现的：
本发明实施例提供了一种多链接透明互连 TRILL分发树故障的通告方 法， 该方法包括：
当组内的成员路由网桥 RB检测到多链接透明互连 TRILL分发树上发 生故障时，所述成员 RB将相应的故障信息通知给所述成员 RB所在组内的 其他成员 RB。
该方法还包括：
所述组内收到所述故障信息的所述其他成员 RB根据接收的故障信息 为所述 TRILL分发树重新选择指定转发者 AF。
通过以下方式为所述 TRILL分发 4对重新选择 AF: 从组内可用的成员 RB中随机选择或组内的可用成员 RB协商从所述组内的可用成员 RB中选 择。
所述组内的成员路由网桥 RB检测到 TRILL分发树上发生故障，包括： 所述 TRILL分发树的一个或多个 ingress RB分别沿所述 TRILL分发树发送 组播连通性检测才艮文，当组内的成员 RB接收不到相应 ingress RB的所述组 播连通性检测报文时，判定相应 ingress RB到所述成员 RB的报文转发出现 故障。
所述故障信息包括： 检测到故障的 RB标识、 所述检测到故障的 RB所 属组的标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间 存在报文转发故障的 ingress RB的标识。
所述组为虚拟路由网桥 RBv， 相应的， 所述 RB所属组的标识为所述 RBv的别名 Nickname。
本发明实施例还提供了一种多链接透明互连 TRILL分发树故障的通告 装置， 适用于组内的成员路由网桥 RB， 包括：
故障检测模块， 配置为检测 TRILL分发树上的故障；
故障通知模块， 配置为在检测到 TRILL分发树上发生故障时， 将相应 的故障信息通知给所述成员 RB所在组内的其他成员 RB。
该装置还包括：
指定转发者 AF选择模块， 配置为根据接收的故障信息为 TRILL所述 分发树重新选择 AF。
AF选择模块进一步配置为通过以下方式为所述 TRILL分发树重新选 择 AF: 从组内可用的成员 RB中随机选择或协商从组内的可用成员 RB中 选择。
所述故障检测模块， 还配置为在接收不到所述 TRILL分发树上的相应 ingress RB发送的所述组播连通性检测报文时， 判定所述相应 ingress RB到 所述成员 RB的报文转发出现故障。
所述故障信息包括： 检测到故障的 RB标识、 所述检测到故障的 RB所 属组的标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间 存在报文转发故障的 ingress RB的标识。
所述组为虚拟路由网桥 RBv, 相应的， RB所属组的标识为所述 RBv 的另1 J名 Nickname
本发明实施例所提供的一种 TRILL分发树故障的通告方法和装置， 能 够实现在 RBv场景下成员 RB之间通告 TRILL分发树故障信息。 本发明实 施例中， 成员 RB检测到分发树上故障， 导致某 ingress沿该分发树的组播 流转发出现问题时， 将故障信息告知组内的成员 RB， 而不需要洪泛到所有 的 RB，并针对该 ingress和该分发树在剩余可用的成员 RB之间再重新选举 出一个新的指定转发者， 而不需要所有使用到该分发树的任意 ingress转发 的组播流都通过新的指定转发者转发到组播接收者。 附图说明
图 1为本发明实施例的一种 TRILL分发树故障的通告方法的流程示意 图；
图 2为本发明实施例在 LAG场景下一个 RB属于一个 RBv的场景示意 图；
图 3为本发明实施例在 LAG场景下一个 RB属于两个 RBv的场景示意 图；
图 4为本发明实施例在 LAN场景的示意图；
图 5为本发明实施例的一种 TRILL分发树故障的通告装置的组成结构 示意图。 具体实施方式
下面结合附图和具体实施例对本发明实施例的技术方案进一步详细阐 述。 本发明实施例提供的一种 TRILL分发树故障的通告方法，如图 1所示， 主要包括：
步骤 101， 当组内的成员 RB检测到 TRILL分发树上发生故障时， 所 述成员 RB将相应的故障信息通知给所述成员 RB所在组内的其他成员 RB。
较佳的， 该方法还可以包括：
步骤 102， 组内收到所述故障信息的其他成员 RB， 根据接收的故障信 息为所述 TRILL分发树重新选择 AF。 优选地： 从组内可用的成员 RB中随 机选择或组内的可用成员 RB协商从所述组内的可用成员 RB中选择。
较佳的，该方法还包括： TRILL分发树的 ingress RB分别沿所述 TRILL 分发树发送组播连通性检测报文， 当所述组内的成员 RB 接收不到相应 ingress RB的所述组播连通性检测报文时， 判定相应 ingress RB到所述成员 RB的报文转发出现故障。
所述故障信息包括： 检测到故障的 RB标识、 所述检测到故障的 RB所 属组的标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间 存在报文转发故障的 ingress RB的标识。 其中， 在 RBv的场景下， 所述组 为 RBv， 相应的， RB所属组的标识为所述 RBv的 Nickname; 在没有 RBv 的场景下， 所述 RB所属组的标识为能唯一标识所述组的标识。
下面结合具体实施例对上述 TRILL分发树故障的通告方法进一步详细 阐述。
首先，如图 2所示，图 2中只有 RB4、RB6、RB11和 RB12是 ingress RB， 针对以 RB10为树根的分发树， RB4、 RB6、 RBI 1和 RBI 2分别发送组播连 通性检测报文检测分发树的连通性， 其中， 各个 ingress RB发出的组播连 通性检测报文的 TRILL头部的 ingress nickname分别是各 ingress RB 的 nickname , egress nickname 是 RBI 0的 nickname; ^口图 2所示的粗实线所 示的分发树， RB3是该分发树的下行指定转发者、 RB1、 RB2和 RB3属于 一个组 RBv 1。 本发明实施例一针对图 2所示的一个成员 RB只属于一个 RBv的情形， 其执行的 TRILL分发树故障的通告方法主要包括：
步骤一， RB4、 RB6、 RBl l和 RB12周期性发送组播连通性检测消息， 检测该分发树的连通性。
步骤二， 假设 RB8设备发生了故障， 则 RB1和 RB2在一定时间内接 收不到 RB11和 RB12发送的组播连通性检测消息， RB3在一定时间内接收 不到 RB4、 RB6、 RBl l和 RB12发送的组播连通性检测消息。
步骤三， RB1、 RB2和 RB3分别发送故障通知消息给 RBvl 内其它的 成员 RB。
这里， RB1发送的故障消息携带了 {RB1， RBvl， RB10， RBl l , RB12} , 故障消息中携带的 RBI表示 RBI是检测到分发树故障的节点， RBvl表示 RB1属于 RBvl的成员 RB， RB10表示树根， RB11和 RB12表示 RB11和 RB12在选择 RB10为树根的分发树向 RB1的转发出现了故障；这里可以通 过 ESADI来承载上述故障信息，且 ESADI帧是可以通过单播的方式分别发 送给 RB2和 RB3的。 RB2发送的故障消息携带了 {RB2， RBvl , RB10, RBl l , RB12} , 故障消息中携带的 RB2表示 RB2是检测到分发树故障的节 点， RBvl表示 RB2属于 RBvl的成员 RB， RB10表示树根， RB11和 RB12 这里也可以通过 ESADI来承载上述故障信息，且 ESADI帧是可以通过单播 的方式分别发送给 RB1和 RB3的。 RB3发送的故障消息携带了 {RB3，RBvl， RB10, RB4, RB6, RBl l , RB12}， 故障消息中携带的 RB3表示 RB3是检 测到分发树故障的节点， RBvl表示 RB3属于 RBvl的成员 RB， RB10表 示树根， RB4、 RB6、 RBl l和 RB12表示 RB4、 RB6、 RBl l和 RB12在选 择 RB 10为树根的分发树向 RB3的转发出现了故障；这里也可以通过 ESADI 来承载上述故障信息， 且 ESADI帧是可以通过单播的方式分别发送给 RB1 和 RB2的。 在 RB 10为根的分发树内重新选择一个指定转发者，具体的：对于 RB4、 RB6在 RBvl组内可用的成员 RB有 RB1和 RB2 , 对于 RB11和 RB12在 RBvl组内则没有可用的成员 RB， 那么该实施例中只需要针对 RB4、 RB6 在 RBvl组内可用的成员 RB: RB1和 RB2中重新选择一个 AF即可。 实际 应用中， 可以采用随机的方式重新选择指定转发者， 也可以采用组内的成 员 RB协商的方式重新选择指定转发者，还可以采用特定的算法来重新选择 指定转发者。
本发明实施例二也针对图 2所示的一个成员 RB只属于一个 RBv的情 形， 其执行的 TRILL分发树故障的通告方法主要包括：
步骤一， RB4、 RB6、 RB11和 RB12周期性发送组播连通性检测消息， 检测该分发树的连通性。
步骤二， 假设 RB10-RB8的链路发生了故障， 则 RB3在一定时间内接 收不到 RB4和 RB6发送的组播连通性检测消息，而可以收到 RBI 1和 RB12 发送的沿该分发树发送的组播连通性检测报文； RB1和 RB2在一定时间内 接收不到 RB 11和 RB 12发送的组播连通性检测消息， 而可以接收到 RB4 和 RB6发送的组播连通性检测消息。
步骤三， RB1、 RB2和 RB3分别发送故障通知消息给 RBvl 内其它的 成员 RB。
这里， RB3发送的故障信息携带了 {RB3， RBvl , RBIO, RB4， RB6}， RBIO说明是树根为 RBIO的分发树上出现故障， RB4和 RB6代表发生的故 障导致哪些 ingress RB发送的组播报文无法被转发到指定的接收者 RB3 ; 该故障信息通过单播 ESADI方式分别通告给 RBvl 内的组内其它成员、 即 RB1和 RB2， RBI和 RB2收到该故障信息后，发现 RB3是 RBvl组内的树 根为 RB10的分发树的指定转发者， 则需要在 RB1和 RB2间重新选出该分 发树在 RB4和 RB6作为 ingress导入组播流时在 RBvl内的唯一的指定转发 者。 假设 RB3不为指定转发者， RBI为该分发树在 RBvl内的指定转发者， RB3 检测到所述分发树上到本节点的分支发生故障， 则通告的信息发送出 来， RB1和 RB2收到之后， 发现 RB3不是指定转发者， 则无需重新进行选 举， 继续沿 RB1转发该分发树的组播流量， 也不影响现有的指定转发者的 正常转发。 实际应用中， 可以采用随机的方式重新选择指定转发者， 也可 以采用组内的成员 RB协商的方式重新选择指定转发者，还可以采用特定的 算法来重新选择指定转发者。
同时， RB1和 RB2发出的故障信息携带了 {RB1， RBvl , RB10, RB11 , RB12}和 {RB2， RBvl , RB10, RB11 , RB12} , 当 RB3收到后， RB3被选 举为 ingress为 RB11和 RB12沿 RB10为根的分发树转发组播流时的指定转 发者。
可以看出， 对于 RB10为根的分发树， 针对 RB4和 RB6沿 RB10为根 的分发树转发组播流时的指定转发者是 RB1和 RB2中的一个； 针对 RB11 和 RB12沿 RB10为根的分发树转发组播流时的指定转发者为 RB3。
在该实施例中，如果还是按照现有的 PN草案的通告方式， 那么所有的 成员都发出故障信息， 则没有可用的指定转发者可以选举， 所有的组播流 在网络收敛之前都被丟弃了。 而按照本发明实施例， 即使所有的成员 RB都 发出故障信息，但是成员在判断出其还可以接收到某些 ingressRB沿该树转 发的组播流时， 为该 ingress RB和该分发树选举新的指定转发者， 在这种 情况下可以保证有可用的成员将接收到的组播流转发至连接的组播接收者。
本发明实施例三针对图 3所示的一个成员 RB属于两个 RBv的情形， 如图 3所示的 CE4发出的组播流经由 RB4作为 ingress RB进入 TRILL网络， RB4选择了以 RB10作为树根的一棵分发树转发该组播流， 如图 3所示的 粗实线所示的分发 f对。 RB2同时属于两个 RBv， RBI和 RB2构成 RBv2， RB2和 RB3构成 RBvl， RB2同时被选举或配置为分发树在 RBvl和 RBv2 内的指定转发者。 TRILL分发树故障的通告方法主要包括： 步骤一， RB4 周期性发送组播连通性检测消息， 检测该分发树的连通 性。
步骤二， RB7-RB2 的链路发生故障， 则 RB2 在一定时间内接收不到 RB4发送的组播连通性检测消息。
步骤三， RB2发送故障通知消息给 RBvl和 RBv2内的其它成员 RB。 这里， RB2发送的消息是两个，一个是发给 RB 1的，携带的信息为 {RB2， RB10, RBv2} , RBI收到该消息后， 发现自己是 RBv2 内唯一可用的成员 RB, 则将自身选举为分发树在 RBv2 内的指定转发者； 另外一个消息是发 送给 RB3的， 携带的信息为 {RB2， RB10, RBvl } , RB3收到该消息后， 发 现自己是 RBvl内唯一可用的成员 RB， 则将自身选举为分发树在 RBvl 内 的指定转发者。
本发明实施例四针对图 4所示的 LAN场景， 如图 4所示， RB1、 RB2 和 RB3分别是该 LAN上 VLAN1-100、 101-200和 201-300的 AF。 TRILL 分发树故障的通告方法主要包括：
步骤一， RB4 周期性发送组播连通性检测消息， 检测该分发树的连通 性； 其中组播 BFD报文中携带的是 VLAN 100， 指明检查的是 VLAN 100的 剪枝树的连通性。
步骤二， RB7-RB1 的链路发生故障， 则 RB1 在一定时间内接收不到 RB4发送的组播连通性检测消息。
这里， VLAN1-99的组播 BFD检查报文传输的组播树不是如图 4中实 线所示的组播树， 当 RB7-RB1的链路发生故障时， RB1只有 VLAN 100的 组播 BFD报文收不到。
步骤三， RB1发送故障通知消息给所述 LAN上的其它成员 RB。
这里的故障信息通告可以是通过组播 ESADI在该 LAN上发送或者是 中间系统-中间系统的会话（IIH， IS-IS HELLO )发送的， 消息中携带的信 息是 {VLAN100}。所述 LAN上的其他成员收到 RB1发出的该故障信息后， 知道 RBI是该 VLAN的 AF， 则给 VLAN100再重新分配一个新的 AF。 该实施例只是在 LAN场景下的一个例子， 如果组播连通性检查报文中 未携带 VLAN信息或者是携带了默认 VLAN, 那么收不到该组播连通性检 查报文的 RB在发送的故障信息中不需要携带 VLAN信息，这样，由于 LAN 上的各个成员是知道所有 VLAN的 AF信息的， 当其他成员收到该故障信 息后，就将以该 RB作为 AF的所有的 VLAN再重新分配新的 RB作为新的 AF (即判断出有哪些 VLAN是以所述 RB作为 AF的， 并为这些 VLAN在 剩余的成员 RB内重新分配新的 AF )。
对应本发明实施例的 TRILL分发树故障的通告方法， 本发明实施例还 提供了一种 TRILL分发树故障的通告装置，该装置适用于组内的成员 RB， 如图 5所示， 该装置主要包括： 故障检测模块 10和故障通知模块 20;
其中， 故障检测模块 10， 配置为检测 TRILL分发树上的故障； 故障通知模块 20， 配置为在检测到 TRILL分发树上发生故障时， 将相 应的故障信息通知给所述成员 RB所在组内的其他成员 RB。
较佳的， 该装置还包括： AF选择模块 30， 配置为根据接收的故障信息 为所述分发树重新选择 AF。
较佳的， AF选择模块 30进一步配置为通过以下方式为所述 TRILL分 发树重新选择 AF: 从组内可用的成员 RB中随机选择或协商从组内的可用 成员 RB中选择。
较佳的， 所述故障检测模块 10， 还配置为在接收不到所述 TRILL分发 树上的相应 ingress RB发送的组播连通性检测报文时，判定所述相应 ingress RB到所述成员 RB的报文转发出现故障。
较佳的， 所述故障信息包括： 检测到故障的 RB标识、 所述检测到故障 的 RB所属组的标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间存在报文转发故障的 ingress RB的标识。
较佳的， 所述组为 RBv， 相应的， 所述 RB所属组的标识为所述 RBv 的 Nickname
另外， 所述故障检测模块、 AF选择模块可由所述成员 RB中的中央处 理器（ CPU， Central Processing Unit )、 处理器（ MPU, Micro Processing Unit )、 数字信号处理器（ DSP, Digital Signal Processor )或可编程逻辑阵列（ FPGA， Field - Programmable Gate Array ) 实现；
所述故障通知模块可由所述成员 RB中具备通信功能的硬件实现。
以上所述， 仅为本发明的较佳实施例而已， 并非用于限定本发明的保 护范围。

Claims

权利要求书
1、一种多链接透明互连 TRILL分发树故障的通告方法，该方法包括： 当组内的成员路由网桥 RB检测到多链接透明互连 TRILL分发树上 发生故障时，所述成员 RB将相应的故障信息通知给所述成员 RB所在组 内的其他成员 RB。
2、 根据权利要求 1所述 TRILL分发树故障的通告方法， 其中， 该方 法还包括：
组内收到所述故障信息的所述其他成员 RB根据接收的故障信息为 所述 TRILL分发树重新选择指定转发者 AF。
3、 根据权利要求 2所述 TRILL分发树故障的通告方法， 其中， 通过 以下方式为所述 TRILL分发树重新选择 AF:从组内可用的成员 RB中随 机选择或组内的可用成员 RB协商从所述组内的可用成员 RB中选择。
4、 根据权利要求 1、 2或 3所述 TRILL分发树故障的通告方法， 其 中， 所述组内的成员路由网桥 RB检测到 TRILL分发树上发生故障， 包 括：
所述 TRILL分发树的一个或多个 ingress RB分别沿所述 TRILL分发 树发送组播连通性检测报文， 当组内的成员 RB接收不到相应 ingress RB 的所述组播连通性检测报文时，判定相应 ingress RB到所述成员 RB的报 文转发出现故障。
5、 根据权利要求 4所述 TRILL分发树故障的通告方法， 其中， 所述 故障信息包括： 检测到故障的 RB标识、 所述检测到故障的 RB所属组的 标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间存在 才艮文转发故障的 ingress RB的标识。
6、 根据权利要求 5所述 TRILL分发树故障的通告方法， 其中， 所述 组为虚拟路由网桥 RBv， 相应的， 所述 RB所属组的标识为所述 RBv的 另 'J名 Nickname
7、 一种多链接透明互连 TRILL分发树故障的通告装置， 适用于组内 的成员路由网桥 RB， 包括：
故障检测模块， 配置为检测 TRILL分发树上的故障；
故障通知模块， 配置为在检测到 TRILL分发树上发生故障时， 将相 应的故障信息通知给所述成员 RB所在组内的其他成员 RB。
8、 根据权利要求 7所述 TRILL分发树故障的通告装置， 其中， 该装 置还包括：
指定转发者 AF选择模块，配置为根据接收的故障信息为所述 TRILL 分发树重新选择 AF。
9、 根据权利要求 8所述 TRILL分发树故障的通告装置， 其中， AF 选择模块进一步配置为通过以下方式为所述 TRILL分发树重新选择 AF: 从组内可用的成员 RB中随机选择或协商从组内的可用成员 RB中选择。
10、 根据权利要求 7、 8或 9所述 TRILL分发树故障的通告装置， 其 中， 所述故障检测模块， 还配置为在接收不到所述 TRILL分发树上的相 应 ingress RB发送的组播连通性检测 文时，判定所述相应 ingress RB到 所述成员 RB的报文转发出现故障。
11、 根据权利要求 10所述 TRILL分发树故障的通告装置， 其中， 所 述故障信息包括： 检测到故障的 RB标识、 所述检测到故障的 RB所属组 的标识、 发生故障的 TRILL分发树标识以及与检测到故障的 RB之间存 在报文转发故障的 ingress RB的标识。
12、根据权利要求 11所述 TRILL分发树故障的通告装置， 其中， 所 述组为虚拟路由网桥 RBv， 相应的， RB所属组的标识为所述 RBv的别 名 Nickname
PCT/CN2013/083916 2013-01-31 2013-09-22 一种trill分发树故障的通告方法和装置 WO2014117530A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310038904.9 2013-01-31
CN201310038904.9A CN103973471B (zh) 2013-01-31 2013-01-31 一种trill分发树故障的通告方法和装置

Publications (1)

Publication Number Publication Date
WO2014117530A1 true WO2014117530A1 (zh) 2014-08-07

Family

ID=51242539

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/083916 WO2014117530A1 (zh) 2013-01-31 2013-09-22 一种trill分发树故障的通告方法和装置

Country Status (2)

Country Link
CN (1) CN103973471B (zh)
WO (1) WO2014117530A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105515999B (zh) * 2014-09-24 2020-05-19 中兴通讯股份有限公司 端系统地址分发信息协议的快速收敛方法及装置
CN104301153B (zh) * 2014-10-31 2018-11-27 新华三技术有限公司 一种trill网络中avf的分配方法、rb设备和drb设备
CN106941446A (zh) * 2016-01-04 2017-07-11 中兴通讯股份有限公司 本地聚合链路处理方法及装置
CN108449276B (zh) * 2018-03-23 2021-01-26 新华三技术有限公司 路由收敛方法及装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110194403A1 (en) * 2010-02-05 2011-08-11 Cisco Technology, Inc. Fault isolation in trill networks
CN102404216A (zh) * 2011-11-23 2012-04-04 华为技术有限公司 一种trill网络保护的方法、路由桥和系统
CN102638389A (zh) * 2011-02-15 2012-08-15 中兴通讯股份有限公司 一种trill网络的冗余备份方法及系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9497073B2 (en) * 2011-06-17 2016-11-15 International Business Machines Corporation Distributed link aggregation group (LAG) for a layer 2 fabric
WO2012106915A1 (zh) * 2011-07-22 2012-08-16 华为技术有限公司 故障通告方法、检测装置、转发装置、系统及数据结构
CN102315951B (zh) * 2011-09-19 2015-05-27 华为技术有限公司 一种组播报文的传输方法、相关设备及系统
CN102970231B (zh) * 2012-11-20 2018-05-01 中兴通讯股份有限公司 组播流转发实现方法和路由网桥（rb）

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110194403A1 (en) * 2010-02-05 2011-08-11 Cisco Technology, Inc. Fault isolation in trill networks
CN102638389A (zh) * 2011-02-15 2012-08-15 中兴通讯股份有限公司 一种trill网络的冗余备份方法及系统
CN102404216A (zh) * 2011-11-23 2012-04-04 华为技术有限公司 一种trill网络保护的方法、路由桥和系统

Also Published As

Publication number Publication date
CN103973471B (zh) 2018-11-02
CN103973471A (zh) 2014-08-06

Similar Documents

Publication Publication Date Title
US20200296025A1 (en) Route Processing Method and Apparatus, and Data Transmission Method and Apparatus
JP5484590B2 (ja) 擬似ワイヤに基づいてサービストラヒックを処理するための方法、デバイスおよびシステム
US7848225B2 (en) Service protecting switching method, system and device in a connectionless network
US9197583B2 (en) Signaling of attachment circuit status and automatic discovery of inter-chassis communication peers
US9509522B2 (en) Forwarding multicast data packets
US8059549B2 (en) Method and apparatus for supporting network communications using point-to-point and point-to-multipoint protocols
US8953590B1 (en) Layer two virtual private network having control plane address learning supporting multi-homed customer networks
CN105991432A (zh) 提供商边缘路由器及方法
CN102035729B (zh) 一种组播数据转发方法
US8165031B2 (en) Multi-point and rooted multi-point protection switching
US9755957B2 (en) Pseudowire control channel for signaling events
CN108512739A (zh) 以太网虚拟专用网络中的多宿主路由器之间的组播状态
US8902794B2 (en) System and method for providing N-way link-state routing redundancy without peer links in a network environment
KR101613640B1 (ko) 패킷 포워딩
WO2013182059A1 (zh) 多协议标签交换流量工程隧道建立方法及设备
EP2890060B1 (en) Vpn implementation method and pe device
CN104767680B (zh) 路由快速切换的方法和装置
WO2014079246A1 (zh) 组播流转发实现方法和路由网桥（rb）
WO2012106915A1 (zh) 故障通告方法、检测装置、转发装置、系统及数据结构
WO2018014767A1 (zh) 一种信息确定方法、装置及存储介质
WO2022021818A1 (zh) 数据报文的处理方法及装置、存储介质、电子装置
WO2011144088A2 (zh) 一种业务保护方法及接入设备
US20130077473A1 (en) Protection switching method and system for a multi-rooted point-to- multi-point service in a provider backbone bridge (pbb) network
WO2019062268A1 (zh) 传输组播报文的方法、装置和系统
WO2014117530A1 (zh) 一种trill分发树故障的通告方法和装置

Legal Events

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

Ref document number: 13874091

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: 13874091

Country of ref document: EP

Kind code of ref document: A1