WO2006034629A1 - Procede de rearrangement de protection dans un systeme mpls - Google Patents

Procede de rearrangement de protection dans un systeme mpls Download PDF

Info

Publication number
WO2006034629A1
WO2006034629A1 PCT/CN2005/001532 CN2005001532W WO2006034629A1 WO 2006034629 A1 WO2006034629 A1 WO 2006034629A1 CN 2005001532 W CN2005001532 W CN 2005001532W WO 2006034629 A1 WO2006034629 A1 WO 2006034629A1
Authority
WO
WIPO (PCT)
Prior art keywords
fault
label
protection switching
node
lsp
Prior art date
Application number
PCT/CN2005/001532
Other languages
English (en)
French (fr)
Inventor
Yuxiang Wang
Ming Yan
Original Assignee
Huawei Technologies Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co., Ltd. filed Critical Huawei Technologies Co., Ltd.
Publication of WO2006034629A1 publication Critical patent/WO2006034629A1/zh

Links

Classifications

    • 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
    • H04L69/40Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection

Definitions

  • the present invention relates to a multi-protocol label switching (MPLS) technology, and in particular, to a protection switching method in an MPLS system.
  • MPLS multi-protocol label switching
  • MPLS Multi-Protocol Label Switching
  • VPI Virtual Path Identifier
  • AV virtual transfer mode
  • LSP label switching paths
  • Label is a short, easy-to-handle information content that does not contain topological information and only has local meaning. Label is short for easy processing and can usually be used. The index method is directly referenced; Label has only local meaning for the ease of distribution of Label.
  • the encapsulation format of the MPLS frame is the Martini encapsulation protocol defined by the Internet Engineering Task Force (IETF) in the draft draft-martini-12circuit-encap-mpls-04.
  • the MPLS frame encapsulated by the Martini encapsulation protocol includes two layers of labels: the outer layer is a tunnel label and the inner layer is a virtual label (VC label, Virtual circuit label).
  • Label Labels can be nested infinitely in MPLS frames, enabling MPLS technology to provide unlimited service support, which is the biggest advantage of MPLS technology.
  • the forwarding path of the service data is determined by the corresponding Label Switched Path (LSP).
  • the label switched path LSP refers to multiple label switching routers (LSR, Label Switch) at a certain logical level. Router) and Label Edge Router (LER, Label Edge Router); the label edge router LER will be equivalence class according to certain rules (FEC, Forwarding Equivalence) Class ), labeling and encapsulating the label of the data packets that are connected to the network system. All label switching routers LSR '3 ⁇ 4 ⁇ in the subsequent MPLS network system are based on the short label Label pair encapsulated in the data packet. The packet is forwarded. When the packet finally leaves the MPLS network system, the label Label encapsulated in the packet will be stripped off by the label edge router LER, and then the LER will strip the Label packet and send it to the next target device.
  • LSR Label Switched Path
  • the tunnel LSP 1 is a tunnel LSP between nodes R1 and R5, and the label switching path is 10 ( R1—>R6) ->20 (R6->R5), where node R6 only completes the tunnel label exchange;
  • VC LSP2 is a virtual channel LSP between nodes R1 to R5, and its label switching path 10.01 ( R1 - > R6 ) - > 20.01 ( R6 -> R5 ), since the node R6 only completes the tunnel label exchange, the inner virtual channel label (VC Label ) is not for the node R6.
  • the packet label encapsulated by the node R1 is 10, and the encapsulated VC Label is 01 (that is, corresponding to the outgoing label 10.01 in the R1 table).
  • the output packet is encapsulated.
  • the Tunnel Label is 20, and the VC Label of the package is still 01 (that is, corresponding to the outgoing label 20.01 in the R6 table).
  • the Telecommunication Standardization Sector of ITU defines the Operation and Maintenance (OAM) function for the MPLS network system.
  • OAM data frame format is defined by ITU-T Rec. Y.1711.
  • OAM frames include: Connectivity Verification (CV) frames, Forward Defect Indication (FDI) frames, Backward Defect Indication (BDI) frames, performance messages, loopback clearing frames, and loopback response frames, but only specific formats and operating procedures are defined for CV, FDI, and BDI frames, where:
  • the CV frame is generated in the source LSR of the LSP, transmitted at a rate of one frame per second, and
  • the LSP is terminated in the sink LSR.
  • the CV frame carries the Trail Termination Source Identifier (TTSI), which provides the basis for detecting all faults.
  • TTSI Trail Termination Source Identifier
  • FFD Fast Failure Detection
  • the function of the FFD frame is defined as an option in ITU-T Rec. Y.1711, that is, in the process of specifically transmitting a frame, it may be set to transmit only the CV frame without transmitting the FFD frame; Or set to send only FFD frames without sending CV frames; FFD frames have multiple transmit speeds that can be set, the maximum transmission speed is up to 100/sec, and the minimum transmission speed is 2/sec.
  • the role of the FDI frame is to respond to the detection of fault behavior (for example, to respond to faults from CV frames).
  • the main purpose of the FDI frame is to suppress the alarm of the network layer above the detected faulty layer.
  • the transmission frequency of the FDI frame is 1. / sec.
  • the BDI frame is inserted on the return path (such as a return LSP) to notify the uplink LSR node (that is, the source node of the forward LSP) the fault defect detected in the sink LSR node of the downlink LSP.
  • the return path such as a return LSP
  • a protection switching mechanism can be implemented in the MPLS system to prevent loss of data packets due to network failure, so as to improve the communication performance of the MPLS system.
  • the following is a description of the 1+1 and 1:1 protection switching mechanisms in the MPLS system (other types of switching mechanisms are similar):
  • the 1+1 and 1:1 protection switching mechanism in the MPLS system is an end-to-end protection technology.
  • a protection LSP is pre-configured for each working LSP in the MPLS system to provide a protection path for the transmission data.
  • the sink node uses the OAM function in the MPLS system to detect the validity of the LSP. When the sink node detects that the LSP is faulty, it starts the protection switching behavior and switches to the protection LSP.
  • FIG. 2 is a schematic diagram of a switching principle of a 1+1 protection switching mechanism in an existing MPLS network system.
  • a data label service is performed by a source label edge router LER of an LSP.
  • the dual-issue that is, the data service is simultaneously sent on the working LSP and the protection LSP, and the LSP sink label edge router LER completes the reception of the data service on the working LSP and the protection LSP (as shown in (a) before the protection state).
  • the OLT performs the validity detection on the working LSP and the protection LSP by using the OAM function in the MPLS system.
  • the sink LER chooses to receive the data service from the protection LSP (as shown in Figure 2). (b) as shown in the post-protection state).
  • FIG. 3 is a switch of a 1:1 protection switching mechanism in an existing MPLS network system.
  • the source label edge router LER of the LSP needs to determine whether the data service is sent through the working LSP or the protection LSP according to the network working condition: that is, when the network working state is normal
  • the data service is sent through the working LSP, and the sink label edge router LER receives the data service from the working LSP correspondingly (as shown in (a) the pre-protection state in FIG. 3); the sink LER detects the working LSP through the OAM function of the MPLS network system.
  • the LER can detect that the working LSP is faulty, switch to the protected LSP to receive data services, and notify the source LER through the reverse LSP to switch the data service.
  • the source LER receives the fault.
  • the data service is switched to the protection LSP for transmission, and the protection of the data service transmission is completed (as shown in (b) protected state in FIG. 3).
  • the connection validity detection of the LSP is performed by the LSP sink node using the OAM function of the MPLS system.
  • the connection validity detection of the LSP can be completed by the CV frame, and the CV frame is transmitted by the source node of the LSP at a rate of 1/sec, and the sink node receives according to the CV frame.
  • the situation is to detect the validity of the connection of the LSP. If the correct CV frame is not received for three consecutive cycles (the period here refers to the time interval for transmitting an OAM frame), the network system is considered to have a connectivity check loss defect (LOCV, Loss of Connectivity Verification)
  • LOCV Loss of Connectivity Verification
  • the transmission rate of the CV frame is 1/sec
  • the three cycles take 3 seconds, that is, the LSP sink node uses the CV frame for LOCV fault detection, and as the start condition of the protection switching, it takes at least 3 seconds.
  • the delay this delay length can not meet the requirements of the protection switching process within 50ms of the carrier class.
  • ITU-T Rec. Y.1711 proposes an alternative improvement, using the improved FFD frame to detect the continuous validity of the LSP.
  • the transmission speed of the FFD frame is higher than the transmission speed of the CV frame, and the transmission speed is optional, specifically 100 / sec, 50 / sec, 20 / sec, 10 / sec, 5 / sec and 2 / sec several transmission speeds. In this way, the time for detecting the LOCV fault by using the FFD frame will be shortened, such as the source end of the LSP.
  • the sink node detects that the time for not receiving the FFD frame correctly for 3 consecutive cycles is only 30 ms, that is, the LOCV fault can be detected in only 30 ms, and the corresponding protection is started.
  • the switching process can shorten the protection switching time in the MPLS system, and lays a foundation for meeting the requirements of completing the protection switching process within 50 ms of the carrier class.
  • the FFD frame can improve the real-time fault detection of the MPLS system, it is only necessary to set the transmission speed of the FFD frame to 100/sec in order to achieve the protection switching processing requirement within 50 ms of the carrier level. However, if the transmission speed of the FFD frame is reduced to 50 / sec, the detection time of the LOCV failure needs 60 ms, which exceeds the carrier-class 50 ms requirement.
  • the OAM frame is inserted into the LSP and transmitted with the service information, which will occupy the service bandwidth of the LSP. Therefore, the OAM frame is not easily transmitted for a long time.
  • the minimum frame length of an FFD frame is 44 bytes, and the encapsulation of two layers of labels (such as Ethernet encapsulation), the minimum frame length of the FFD frame will reach 64 bytes, and if it is to meet the carrier-class requirements, 64 The byte FFD frame is transmitted at a rate of 100/sec.
  • ITU-T Rec. Y.1711 recommends setting the transmission speed of the FFD frame to 20/sec, but this will cause the detection time of the LOCV fault to be as long as 150 ms, that is, the sink node of the LSP needs at least 150 ms delay.
  • the protection switching process is completed, and the protection switching speed is low, which cannot meet the requirements of completing the protection switching process within 50 ms of the carrier class.
  • the technical problem to be solved by the present invention is to propose a protection switching method in a multi-protocol label switching system to improve the protection switching processing speed of the label switching path.
  • the present invention provides a protection switching method in a multi-protocol label switching system, including the steps of:
  • the step (1) also includes the steps before:
  • the carrier-class protection switching time requirement in the step (P1) is that the protection switching process is completed in N milliseconds, and the N is a positive integer; the sending rule is initially greater than or equal to ⁇
  • the transmission speed is reduced, and the alarm indication signal is sent until the fault is recovered.
  • the sink node in step (3) After receiving the m alarm indication signals continuously, the sink node in step (3) starts protection switching processing, where m is a natural number.
  • the case of detecting the fault specifically includes:
  • the node finds that the physical link corresponding to its physical port is invalid; or
  • (21) establishing a mapping relationship table capable of finding the outbound port number, the inbound label, and the outgoing label according to the port number;
  • the above inbound and outbound tags are tunnel tags.
  • the node that finds the fault in the step (2) sends the alarm indication signal from the corresponding egress port according to the egress port number corresponding to the associated label switching path.
  • the port can be a logical port or can be a physical port.
  • the method also includes the steps of:
  • the node that finds the fault sends a fault cancellation signal to the sink node in each associated label switching path.
  • the sink node starts the protection switching recovery process after receiving the fault cancellation signal.
  • the foregoing alarm indication signal is: a forward defect indication frame; or an operation and maintenance message including an operation and maintenance function type, a defect type, and a fault node ID information.
  • the sink node may start the protection switching process after receiving the multiple alarm indication signals. It is also possible to start the protection switching process after receiving a plurality of alarm indication signals continuously.
  • the fault is discovered by any node in the MPLS system, and the discovered fault information is actively diffused to the sink node of each LSP related to the fault, so that the sink node in each related LSP can receive the enemy obstacle information quickly and timely.
  • the protection processing speed of the label switching path can be improved.
  • FIG. 1 is a schematic diagram of a process of forwarding a data packet by a conventional MPLS network system
  • FIG. 2 is a schematic diagram of a switching principle of a 1+1 protection switching mechanism in an existing MPLS network system
  • FIG. 3 is a 1:1 diagram of an existing MPLS network system. Schematic diagram of the switching principle of the protection switching mechanism
  • FIG. 4 is a flow chart showing the main implementation principle of the protection switching method in the multi-inch label switching system of the present invention
  • Figure 5 is a flow chart showing an embodiment of setting a transmission rule in a protection switching method in the multi-protocol label switching system of the present invention
  • FIG. 6 is a schematic diagram of a process of finding a failure of a node in a protection switching method in the multi-protocol label switching system of the present invention, and correspondingly searching for an LSP related to the failure;
  • FIG. 7 is a transmission switching method for transmitting a fault cancellation signal in a multi-protocol label switching system of the present invention. Flow chart of the embodiment of the number;
  • FIG. 8 is a schematic diagram of an implementation process of a protection switching method in a multi-protocol label switching system of the present invention in an MPLS system topology diagram.
  • each LSP in an MPLS system is an end-to-end connection path, specifically a switched packet transmission channel composed of multiple LSRs and LERs at a certain logical level; OAM in the MPLS system -
  • the force-energy feature provides an end-to-end detection management mechanism for each LSP. Based on this detection management mechanism, the sink node on each LSP can switch to the protection LSP and continue working when the working LSP fails. Implement the protection switching mechanism in the MPLS system.
  • the sink node on each LSP is passive for the detection of the fault, and because the OAM frame is not sent frequently for a long time, the detection of the fault at the sink node on each LSP is inevitable to a certain extent.
  • the protection switching method in the multi-protocol label delivery system of the present invention mainly provides that an LSP in an MPLS system finds an LSP related to a fault in time when an enemy is found, and extends the fault information back to each time in time.
  • the LSP on the LSP is related to the LSP. Therefore, the LSP must be passively passed through the CV OAM frame to detect the validity of the LSP. Therefore, the fault information can be quickly and quickly reach the faulty LSP.
  • the sink node that discovers the fault by any node in the MPLS system and proactively spreads the fault to the backend can make the sink node of the faulty LSP in the MPLS system know the fault condition in time. , Perform 'fast protection switching process, which can meet the requirements of carrier-class protection switching processing.
  • FIG. 4 is a flowchart of the main implementation principle of the protection switching method in the multi-protocol label switching system of the present invention.
  • the main implementation process includes:
  • Step S10 The multi-protocol label is exchanged with any node in the MPLS system (including each label switching router LSR or each label edge router LER of each label switching path LSP_t in the MPLS system) to find a fault, and find a fault phase.
  • Off label switching path The case where any node in the MPLS system finds a fault may be as follows: 1) The node finds that the physical link corresponding to its physical port is invalid, such as the case where the Ethernet device finds that its Ethernet port is down (link down);
  • VCG Virtual Concatenatio Group
  • MSTP Multi-Service Transport Platform
  • the node finds its own framing procedure (GFP, General Frame Protocol), and the frame synchronization loss alarm occurs;
  • GFP General Frame Protocol
  • the faulty node finds the label-switched path associated with the fault according to the fault condition.
  • the LSP can be completed in two ways. The first method is as follows:
  • mapping relationship table that can find out the outbound port number, the inbound label, and the outgoing label according to the inbound port number is established in advance; the mapping relationship table established therein may be as follows:
  • the faulty node is determined according to the fault condition, and the inbound port number corresponding to the fault is determined. Then, according to the inbound port number corresponding to the fault, the mapping table stored by the fault is searched, and the corresponding outbound port number, inbound label, and outbound label are obtained by the index. Letter,
  • the label switching path LSP determined by each pair of inbound labels and outbound bids obtained by the above index is used as the label switching path LSP associated with the fault, :3 ⁇ 4.
  • an inbound label label1 and an outgoing label label2 can be used to determine a corresponding LSP.
  • the second method is to directly use the local service forwarding table already stored in each node device in the MPLS system, where the local service forwarding table has recorded information such as the ingress port number, the out port number, the inbound label, and the outgoing label. Correspondence relationship, so that the local service forwarding table can be directly queried according to the inbound port number corresponding to the fault, and the corresponding outbound port number and the corresponding LSP determined by the inbound label and the outgoing label are indexed.
  • the specific process is as follows: The node that finds the fault determines the entry number corresponding to the fault according to the fault condition;
  • the local storage directory is queried: ⁇ service forwarding table, and the corresponding port number, inbound label, and outgoing label information are obtained by the index;
  • ticks switching path LSP determined by each pair of inbound labels and outbound labels obtained as described above is used as the label switching path LSPo associated with the fault.
  • the faulty node is generally only inserted into the LSP determined by the tunnel label to transmit fault information. Therefore, the above-mentioned inbound label and outgoing label are generally Refers to the tunnel label ( Tunnel Label ).
  • the port mentioned above can be a logical port or a physical port.
  • step S20 the node that finds the fault sends an alarm indication signal AIS carrying the fault information to the sink node in each associated label switched path LSP determined in step S10.
  • the alarm indication signal AIS described herein may be
  • the forward defect refers to the frame FDI; it may also include the type of operation and maintenance function (ie, OAM function type), defect type, fault node ID information, Trail Termination Source Identifier (TTSI), signal transmission frequency, etc. Operation and maintenance of messages.
  • the node that finds the fault can exchange the alarm indication signal AIS from the corresponding egress port according to the relevant label switching path LSP. (Refer to the above table for details.)
  • Step S30 After receiving the alarm indication signal AIS, the sink node of each relevant LSP starts the protection switching process.
  • the embodiment uses the operation of initiating the protection switching process after receiving the corresponding number of AISs, that is, the AIS is stably received, and the faulty working LSP is switched to the protection LSP.
  • the sink node on each relevant LSP needs to receive the corresponding number of AISs before performing the protection switching process. The purpose is to remove the jitter. If the working LSP is determined to be faulty, the protection switching process is performed to avoid In the case of erroneous reception of the AIS, the protection switching process is started, and the processing resources of the shovel system are started. In other embodiments, the protection switching process is initiated after successively receiving a corresponding number of AISs.
  • the sink node can discover the fault state of the LSP in time, and perform fast protection switching processing. Ii, further make the processing speed
  • the present invention can further reduce the impact on the service bandwidth of the system, and further proposes that a sending rule can be set in advance, so that the node that finds the fault sends the fault information by using the set sending rule; Referring to FIG. 5, which is a flowchart of an embodiment of setting a transmission rule in a protection switching method in a multi-protocol label switching system according to the present invention;
  • Step S110 pre-setting a transmission rule that can meet the carrier-class protection switching time requirement
  • Step S120 discovering a fault by any node in the multi-protocol label switching MPLS system, and finding a label switching path LSP related to the fault
  • step S130 the node that finds the fault sends an alarm indication signal AIS carrying the fault information to the sink node in each associated label switched path LSP according to the sending rule preset in step S110.
  • Step S140 After receiving the corresponding number of alarm indication signals AIS, the sink node of each associated LSP starts the protection switching process.
  • step S110 when the requirement of the carrier level protection switching time in step S110 is that the protection switching process is completed within N milliseconds (the N is a positive integer, generally 50), and the sink node is continuous in step S140. After receiving m alarm indication signals AIS, the protection switching process is started.
  • the set transmission rule is: Fault
  • the initial state is sent continuously with m alarm indications at a transmission speed greater than or equal to ⁇ /s.
  • the sink node that guarantees the faulty LSP can receive m alarm indication signals in time, and quickly complete the protection switching process, and reduce the transmission rate to send the alarm indication signal AIS (if it can be reduced to 1/sec)
  • the transmission speed sends an alarm indication signal AIS) until the fault is recovered.
  • the sink node of the failed LSP can ensure that the required m AISs are received in time to perform protection switching processing, and the service bandwidth of the system can be minimized.
  • FIG. 6 is a protection switching circuit in the multi-protocol label switching system of the present invention.
  • the label switch router LSR/label edge router LER represents any LSR or LER device on an LSP, if a physical or The physical layer or the data link layer of the logical port is faulty, such as the Ethernet port link down, or a virtual concatenation VC in the virtual concatenation group VCG in the multi-service transport platform MSTP is invalid, or the general framing procedure GFP appears.
  • the frame synchronization loss alarm and the signal failure (SF, Signal fail) state in the service layer of the MPLS system, the LSR/LER device performs the following process by using the solution of the present invention:
  • the LSR/LER finds its own port m port link down
  • the LSR/LER uses the port number of the port m port on which the fault has occurred to query the local service forwarding table stored by itself, and obtains the LSP associated with the disconnected port m, respectively:
  • the LSR/LER inserts AIS information carrying the physical link failure corresponding to the port m on LSP Q, LSP M and LSP N according to a certain transmission rule.
  • the AIS may be a newly defined OAM frame, that is, the above operation and maintenance message including information such as an OAM function type, a defect type, a faulty node ID information, a path source identifier TTSI, and a signal transmission frequency, and may also be a standard.
  • the existing FDI frame has a certain transmission rule:
  • the LSR/LER sends AIS information quickly in the initial stage of the fault, for example, it can send an AIS message every 5 milliseconds or every 10 ms at the beginning of the fault; After the AIS information can be correctly identified by the sink node in the LSP, if five AIS messages or 10 AIS messages are sent continuously, the transmission speed is reduced, for example, it can be reduced to 1/sec until the fault recovery state.
  • Such a transmission rule can ensure that the fault is advertised to the sink node of the LSP in the event of a LSP failure, and the occupation of the effective service bandwidth of the LSP can be reduced.
  • FIG. 7 is a flowchart of an embodiment of a protection switching method for transmitting a fault cancellation signal in a multi-protocol label switching system according to the present invention; the implementation process is as follows:
  • Step S210 discovering a fault by any node in the multi-protocol label switching MPLS system, And finding a label switching path LSP related to the fault;
  • Step S220 The node that finds the fault sends an alarm indication signal AIS carrying the fault information to the sink node in each associated label switched path LSP according to the sending rule preset in step S110.
  • Step S230 After receiving the corresponding number of alarm indication signals AIS, the sink node of each associated LSP starts the protection switching process, that is, the working LSP is switched to the protection LSP to continue working;
  • Step S240 After finding the fault, the node that finds the fault sends a fault cancellation signal to the sink node in each associated label switching path LSP.
  • Step S250 After receiving the fault cancellation signal, the sink node of each associated LSP starts the protection switching recovery process, that is, the protection LSP is switched to the working LSP to continue working.
  • FIG. 8 is a schematic diagram of an implementation process of a protection switching method in a multi-protocol label switching system according to the present invention in a topology diagram of an MPLS system. It is assumed that the LSP Q in FIG. 8 is obtained by LER 1 through LSR 2, LSR 3 An LSP of LER 3, when the MPLS service layer between LSR2 and LER 1 fails, LSR2 obtains the affected port number according to the fault, and queries the local service forwarding table according to the affected port number, and indexes the fault to the fault.
  • LSR 2 will then drop the bearer into the LSP Q path with a label of 40 and an outbound port being a port between LSR 2 and LSR 3.
  • the AIS information with the corresponding fault information such as the FDI frame, etc.; after receiving the corresponding number of AIS information, the LER 3 determines that the corresponding LSP is faulty, and starts the protection switching process of the corresponding LSP.
  • the faulty node is only inserted into the LSP consisting of the outermost label (ie, the Tunnel Label).

Landscapes

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

Description

多协议标签交换系统中的保护倒换方法 技术领域 本发明涉及多协议标签交换( MPLS , Multi protocol Label Switched ) 技术, 尤其涉及一种 MPLS系统中的保护倒换方法。
背景技术 多协议标签交换 ( MPLS, Multi protocol Label Switched )技术兴起于 90年代中期, 其集中了异步传输模式( ATM , asynchronous transfer mode ) 中的虚通道标识(VPI, Virtual Path Identifier ) /虚通路标识(VCI, Virtual Channel Identifier )等一些交换思想, 并无缝的集成了 IP路由技术的灵活 性和二层交换技术的简捷性,从而实现了在无连接的 IP网络中增加 MPLS 面向连接的特性。 在 IP网络系统中, 通过采用 MPLS技术建立多条标签 交换路径(LSP, Label Switched Path ), 来为 IP网络系统提供一些管理和 运营手段。
MPLS技术的关键在于引入了标签(Label ) 的概念, Label是一种短 的、 易于处理的、 不包含有拓朴信息、 只具有局部意义的信息内容, Label 短是为了易于处理, 通常可以用索引方式直接引用; Label只具有局部意 义是为了 Label便于分配。
目前, MPLS帧的封装格式是由 Internet工程任务组(IETF, Internet Engineering Task Force )在草案 draft-martini-12circuit-encap-mpls-04中定义 的 Martini封装协议。 根据 Martini封装协议封装的 MPLS帧, 包含有两层 标签: 其外层为隧道标签(Tunnel label ), 内层为虚电路标签( VC label, Virtual circuit label )。标签 Label在 MPLS帧中可以无限嵌套,从而使 MPLS 技术能够提供无限的业务支持能力, 这正是 MPLS技术的最大优势所在。
在 MPLS网络系统中,业务数据的转发路径是由相应的标签交换路径 ( LSP, Label Switched Path )决定的, 标签交换路径 LSP是指在某逻辑层 次上由多个标签交换路由器(LSR, Label Switch Router )和标签边缘路由 器(LER, Label Edge Router )组成的交换式分组传输通道; 其标签边缘 路由器 LER会根据一定的规则作等价类划分( FEC, Forwarding Equivalence Class ), 对接入到网络系统中的数据包进行标签 Label的分配和封装处理, 后续 MPLS网络系统中的所有标签交换路由器 LSR '¾卩要根据该数据包中 封装的这个简短的标签 Label对数据包进行转发处理, 当该数据包最终要 离开 MPLS网络系统时, 数据包中封装的标签 Label将被标签边缘路由器 LER剥离掉, 然后 LER将剥离掉 Label的数据包发送到下一目标设备。
如图 1所示,该图是现有 MPLS网络系统对数据包进行转发处理的过 程示意图; 图中所示 Tunnel LSP 1为节点 R1至 R5之间的一条隧道 LSP , 其标签交换路径为 10 ( R1—>R6 ) ->20 ( R6->R5 ), 其中节点 R6只完 成隧道标签( Tunnel Label )的交换; 所示 VC LSP2为节点 R1至 R5之间 的一条虚通道 LSP, 其标签交换路径为 10.01 ( R1— >R6 )— >20.01 ( R6 ->R5 ), 由于节点 R6只完成隧道标签(Tunnel Label )的交换, 其内层的 虚通道标签( VC Label )对节点 R6而言是不可见的, 所以由节点 R1发出 的数据包中封装的 Tunnel Label为 10, 封装的 VC Label为 01 (即对应 R1 表格中的出标签 10.01 ), 经节点 R6的标签交换后, 输出数据包中封装的 Tunnel Label为 20, 封装的 VC Label仍为 01 (即对应 R6表格中的出标签 20.01:)。上述过程即为 MPLS网络系统对数据包进行转发处理的简要过程。
为了更好的对 MPLS 网络系统进行维护和管理, 国际电信联盟 -电信 标准化组织 ( ITU T, Telecommunication Standardization Sector of ITU )为 MPLS网络系统定义了操作和维护(OAM, Operation and Maintenance )功 能, 具体 MPLS OAM数据帧格式是由 ITU-T Rec. Y.1711定义的, 目前已 经定义的 OAM帧包括: 连通性校验(CV, Connectivity Verification ) 帧、 前向缺陷指示( FDI , Forward Defect Indication )帧、 后向缺陷指示( BDI , Backward Defect Indication )帧、 性能报文、 环回清求帧和环回响应帧, 但 是只对 CV、 FDI、 BDI三种帧定义了具体的格式和操作规程, 其中: CV帧在 LSP的源 LSR中产生, 以每秒一个帧的速度进行发送, 并在
LSP的宿 LSR中终结; 其中 CV帧携带有网絡的路径源端标识符(TTSI, Trail Termination Source Identifier ) , 这样就能奠定检测所有故障的基础。
在后续的技术发展过程中, 为解决 CV帧发送频率过低, LSP故障检 测不及时的问题, 进而对 CV帧进行了改进, 提出了一种快速故障检测帧 5 001532
-3 -
( FFD , Fast Failure Detection ), 其中 FFD帧的功能在 ITU-T Rec. Y.1711 中定义为可选项, 即在具体发送帧的过程中, 可以设置只发送 CV帧, 而 不发送 FFD帧; 或者设置只发送 FFD帧, 而不发送 CV帧; FFD帧具有 可设置的多个发送速度, 最高发送速度可达 100个 /秒, 最低发送速度为 2 个 /秒。
FDI帧的作用是对检测故障行为的响应 (例如对来自于 CV帧所承载 故障缺陷进行响应), 其主要目的是压制检测到错误的层以上的网络层的 告警, FDI帧的发送频率为 1个 /秒。
BDI帧是在返回通路 (比如一个返回的 LSP )上被插入的, 用来将在 下行 LSP的宿 LSR节点中检测到的故障缺陷通知给上行源 LSR节点(即 前向 LSP的源节点)。
利用上述 MPLS网络系统中的 OAM功能, 可以在 MPLS系统中实现 保护倒换机制, 以防止数据包因网络故障而出现丟失损失, 以提高 MPLS 系统的通信性能。 下面以 MPLS系统中的 1+1和 1:1保护倒换机制来说明 (其他类型的保换倒换机制原理类似):
其中 MPLS系统中的 1+1和 1:1保护倒换机制是一种端到端的保护技 术,通过为 MPLS系统中的每个工作 LSP预先配置一条保护 LSP, 以为传 输数据提供一条保护路径; LSP上的宿节点通过 MPLS系统中的 OAM功 能进行 LSP的有效性检测, 当宿节点检测到 LSP出现故障时, 将启动保 护倒换行为, 切换到保护 LSP上工作。
请参阅图 2, 该图是现有 MPLS网络系统中 1+1保护倒换机制的切换 原理示意图; 在 MPLS系统中, 对于 1+1保护倒换机制, 是由 LSP的源 标签边缘路由器 LER进行数据业务的双发,即将数据业务同时在工作 LSP 和保护 LSP上发送, LSP的宿标签边缘路由器 LER完成在工作 LSP和保 护 LSP上选择接收数据业务(如图 2中 (a )保护前状态所示)。 宿 LER 在工作过程中, 利用 MPLS系统中的 OAM功能对工作 LSP和保护 LSP 分别进行有效性检测, 当检测到工作 LSP出现故障时, 宿 LER选择从保 护 LSP上接收数据业务(如图 2中 ( b )保护后状态所示)。
请参阅图 3, 该图是现有 MPLS网络系统中 1:1保护倒换机制的切换 原理示意图; 在 MPLS系统中, 对于 1:1保护倒换机制, LSP的源标签边 缘路由器 LER需要根据网絡工作情况决定数据业务是通过工作 LSP发送, 还是通过保护 LSP发送: 即当网絡工作状态正常时,将数据业务通过工作 LSP发送,宿标签边缘路由器 LER相应地从工作 LSP上接收数据业务(如 图 3中( a )保护前状态所示); 宿 LER通过 MPLS网络系统的 OAM功能 检测工作 LSP的连接有效性, 当网络出现故障时, 宿 LER可以检测到工 作 LSP出现故障, 及时切换到保护 LSP上接收数据业务, 同时通过反向 LSP通知源 LER进行数据业务的发送切换, 源 LER接收到通知消息后, 将数据业务切换到保护 LSP上进行发送, 完成对数据业务传输的保护(如 图 3中 ( b )保护后状态所示)。
综上所述, 在 MPLS网络系统中, 无论是 1+1保护倒换机制还是 1:1 保护倒换机制, LSP 的连接有效性检测都是由 LSP宿端节点利用 MPLS 系统的 OAM功能来完成的。
在 ITU-T Rec. Y.1711中规定, LSP的连接有效性检测可以通过 CV帧 来完成, CV帧由 LSP的源端节点以 1个 /秒的速度发送, 宿端节点根据 CV帧的接收情况来检测 LSP的连接有效性, 如果连续三个周期 (这里的 周期是指发送一个 OAM帧的时间间隔)没有接收到正确的 CV帧, 则认 为网络系统出现连通性校验丟失缺陷 ( LOCV , Loss of Connectivity Verification )故障, 宿端节点应该上报告警, 并同时启动相应的保护倒换 处理流程。但是由于 CV帧的发送速率为 1个 /秒, 三个周期则需要 3秒的 时间,也就是说 LSP宿端节点使用 CV帧进行 LOCV故障检测,并作为保 护倒换的启动条件, 至少需要 3秒的延时, 这个延时长度根本无法达到电 信级 50ms内完成保护倒换处理的要求。
为了在上述基 上提高保护倒换处理的速度, ITU-T Rec. Y.1711又提 出了一种可选的改进方案, 利用改进的 FFD帧来检测 LSP的连续有效性。 如前所述, 因为 FFD帧与 CV帧相比之下, 其作用基^目同, 唯一不同的 是 FFD帧的发送速度比 CV帧的发送速度高, 且发送速度可选, 具体有 100个 /秒、 50个 /秒、 20个 /秒、 10个 /秒、 5个 /秒和 2个 /秒几个发送速度。 这样利用 FFD帧来检测 LOCV故障的时间将会有所缩短, 如 LSP的源端 节点以 100个 /秒的速度发送 FFD帧, 则宿端节点检测到连续 3个周期没 有正确接收到 FFD帧的时间只需 30ms,即只需 30ms就可以检测到 LOCV 故障, 并启动相应的保护倒换处理, 从而可以缩短 MPLS系统中的保护倒 换时间, 为满足电信级 50ms内完成保护倒换处理的要求奠定了基础。
尽管使用 FFD帧能够提高 MPLS系统的故障检测实时性, 但是要达 到电信級 50ms内完成保护倒换处理的要求, 只有将 FFD帧的发送速度设 置在 100个 /秒才可能满足。 而如果将 FFD帧的发送速度降低到 50个 /秒, 则 LOCV故障的检测时间需要 60ms, 已经超出了电信级 50ms的要求。
同时因为使用 MPLS系统的 OAM功能来检测故障, 是通过将 OAM 帧插入到 LSP中和业务信息一起传输的, 这样必将会占用 LSP的业务带 宽, 所以 OAM帧不易长时间的频繁发送。 例如 FFD帧的最小帧长为 44 字节, 再加上两层标签的封装(如以太网封装), FFD 帧的最小帧长将达 到 64字节, 而如果要达到电信级的要求, 将 64字节的 FFD帧以 100个 / 秒的速度来发送, 则在一个 LSP中, 仅 FFD帧就将占用至少 64 X 8 X 100 = 51200bit/S的带宽, 因此这种保护倒换方式不便于在大型 MPLS网络系 统中进行 OAM功能的推广。由此 ITU-T Rec. Y.1711规定推荐设置 FFD帧 的发送速度为 20个 /秒, 但是这样又将导致 LOCV故障的检测时间长达 150ms, 即 LSP的宿端节点需要至少 150ms延时才能完成保护倒换处理, 保护倒换处理速度低,无法满足电信级 50ms内完成保护倒换处理的要求。
发明内容
本发明要解决的技术问题在于提出一种多协议标签交换系统中的保 护倒换方法, 以提高标签交换路径的保护倒换处理速度。
为解决上述问题, 本发明提出了一种多协议标签交换系统中的保护倒 换方法, 包括步骤:
( 1 ) 由多协议标签交换系统中的任意节点发现故障, 并查找与故障 相关的标签交换路径;
( 2 ) 发现故障的节点分别向每条相关的标签交换路径中的宿端节点 发送承载有故障信息的告警指示信号;
( 3 ) 所述宿端节点在接收到告警指示信号后, 启动保护倒换处理。 所述步骤( 1 )之前还包括步骤:
( P1 )预先设置能够满足电信级保护倒换时间要求的发送规则; 步骤 (2 ) 中发现故障的节点以所述发送规则分别向每条相关的标签 交换路径中的宿端节点发送承载有故障信息的告警指示信号。
其中步骤(P1 )中所述电信级保护倒换时间要求是在 N毫秒内完成保 护倒换处理, 所述 N为正整数; 所述发送规则为初始以大于等于^ ^个
N
/秒的发送速度连续发送 m个告警指示信号后, 降低发送速度发送告警指 示信号至故障恢复时终止;
步骤 (3 ) 中所述宿端节点在连续接收 m个告警指示信号后, 启动保 护倒换处理, 所述 m为自然数。
所述发现故障的情况具体包括:
所述节点发现自身物理端口对应的物理链路失效; 或
发现自身多业务传送平台中的虚级联组失效; 或
发现自身通用成帧规程出现帧同步丢失告警。
戶斤述查找与故障相关的标签交换路径的过程具体包括:
( 21 )建立能够才 据入端口号查找到出端口号、 入标签和出标签的映射 关系表;
( 22 )确定所述故障对应的入端口号;
( 23 )才艮据故障对应的入端口号查询所述映射关系表, 得到对应的出端 口号、 入标签和出标签信息;
( 24 )由得到的每对入标签和出标签分别确定的标签交换路径作为与故 障相关的标签交换路径。
所述查找与故障相关的标签交换路径的过程具体包括:
( 2a )确定所述故障对应的入端口号;
( 2b )根据故障对应的入端口号查询本地业务转发表, 得到对应的出端 口号、 入标签和出标签信息;
( 2c ) 由得到的每对入标签和出标签分别确定的标签交换路径作为与故 障相关的标签交换路径。 上述入标签和出标签为隧道标签。 ― 其中所述步骤(2 ) 中发现故障的节点根据相关的标签交换路径对应 的出端口号将告警指示信号由相应的出端口发出。 所述端口可以为逻辑端 口或可以为物理端口。
所述方法还包括步骤:
( 4 )发现故障的节点在故障消除后, 分别向每条相关的标签交换路径 中的宿端节点发送故障消除信号;
( 5 )所述宿端节点、在接收到故障消除信号后, 启动保护倒换恢复处理。 上述告警指示信号为:前向缺陷指示帧;或包含有操作维护功能类型、 缺陷类型、 故障节点 ID信息的操作维护报文。
所述方法中, 宿端节点可以在接收多个告警指示信号后, 启动保护倒 换处理。 也可以在连续接收多个告警指示信号后, 启动保护倒换处理。 本发明能够达到的有益效果如下:
由 MPLS系统中的任意节点发现故障,并将发现的故障信息主动扩散 到与故障相关的每条 LSP的宿端节点, 使每条相关的 LSP中的宿端节点 能够快速及时接收到敌障信息, 以启动保护倒换处理机制, 能够提高标签 交换路径的保护倒提处理速度。
附图说明
图 1是现有 MPLS网络系统对数据包进行转发处理的过程示意图; 图 2是现有 MPLS网络系统中 1+1保护倒换机制的切换原理示意图; 图 3是现有 MPLS网络系统中 1:1保护倒换机制的切换原理示意图; 图 4是本发明多寸办议标签交换系统中的保护倒换方法的主要实现原理 流程图;
图 5是在本发明多协议标签交换系统中的保护倒换方法中设置发送规 则的实施例流程图;
图 6是在本发明多协议标签交换系统中的保护倒换方法中一个节点发 现故障, 并相应查找与故障相关的 LSP的过程示意图;
图 7是本发明多协议标签交换系统中的保护倒换方法发送故障消除信 号的实施例流程图;
图 8是本发明多协议标签交换系统中的保护倒换方法在 MPLS系统拓 朴图中的实施过程示意图。
具体实施方式
背景技术已经叙述了 MPLS系统中的每条 LSP是一种端到端的连接路 径, 具体是指在某种逻辑层次上曲多个 LSR和 LER组成的交换式分组传 输通道; MPLS系统中的 OAM -力能特性为每条 LSP提供了一套端到端的 检测管理机制,基于这种检测管理机制,每条 LSP上的宿端节点能够在工 作 LSP发生故障时, 及时切换到保护 LSP上继续工作, 实现 MPLS系统 中的保护倒换处理机制。但是由亍每条 LSP上的宿端节点对于故障的检测 是被动的, 同时由于 OAM帧不曷长时间频繁发送, 导致每条 LSP上的宿 端节点对故障的检测不可避免的存在一定程度的延时,从而使 MPLS网络 系统中的保护倒换处理往往由 t这种延时而达不到电信级的保护倒换处 理要求。 本发明多协议标签交提系统中的保护倒换方法主要提供一种由 MPLS系统中的任意节点在发现敌障时,及时查找与发现故障相关的 LSP, 并将故障信息及时向后扩展到每夺相关的 LSP上的宿端节点,从而不需要 在 LSP发生故障时, 必须由 LSP的宿端节点被动地通过 CV OAM帧完成 LSP的有效性检测,因此可以使故障信息及时快速到达发生故障的 LSP的 宿端节点, 通过这种由 MPLS系统中的任意节点发现故障, 并将发现的故 障主动向后延扩散的机制,可以使 MPLS系统中发生故障的 LSP的宿端节 点能够及时得知故障情况, 进行'快速保护倒换处理, 从而能够达到电信级 保护倒换处理的要求。
下面将结合各个附图对本发明多协议标签交换系统中的保护倒换方 法的具体实施过程进行详细的阐述。 请参阅图 4, 该图是本发明多协议标 签交换系统中的保护倒换方法的主要实现原理流程图; 其主要实现过程具 体包括:
步骤 S10, 由多协议标签交 奂 MPLS系统中的任意节点 (包括 MPLS 系统中每条标签交换路径 LSP _t的每个标签交换路由器 LSR或每个标签 边缘路由器 LER )发现故障, 查找与故障相.关的标签交换路径; 其中 MPLS系统中的任意节点发现故障的情况可以 括如下情况: 1 ) 节点发现自身物理端口对应的物理链路失效, 如以太网设备发现 自身的以太网端口断开 (link down ) 的情况;
2 ) 节点发现自身多业务传送平台 ( MSTP , Multi-Service Transport Platform) 中的虚级联组 ( VCG, Virtual Concatenatio Group ) 失效, 如某 虚级联 VC出现告警指示信号 (AIS , Alarm Indication Signal );
3 )节点发现自身的通用成帧规程 ( GFP, General Frame Protocol ) 出 现帧同步丢失告警;
当然节点发现的故障在上述三种故障之外的其他故障也在本发明的 保护范围之内。
其中发现故障的节点根据故障情况查找与故障相关的标签交换路径 LSP可以由两种方式来完成; 其中第一种方式如下:
预先建立能够根据入端口号查找到出端口号、 入标签和出标签的映射 关系表; 其中建立的映射关系表可以如下:
Figure imgf000011_0001
并将建立的映射关系表存储在 MPLS系统中的每个节点设备中;
然后发现故障的节点根据故障情况确定该故障对应的入端口号; 再根据该故障对应的入端口号来查询自身存储的上述映射关系表, 进 而索引得到对应的出端口号、 入标签和出标签信,
最后由上述索引得到的每对入标签和出标答分別确定的标签交换路 径 LSP作为与该故障相关的标签交换路径 LSP, :¾。 >据上表, 由一个入标 签 labell和一个出标签 label2便可以确定一个相应的 LSP。
第二种方式是直接利用 MPLS系统中的每个节点设备中已经存储的本 地业务转发表, 其中本地业务转发表中已经记载有入端口号、 出端口号、 入标签和出标签等信息之间的对应关系, 这样就可以根据故障对应的入端 口号直接查询本地业务转发表, 索引到相应的出端口号和由入标签和出标 签确定的对应 LSP, 具体过程如下: 发现故障的节点根据故障情况确定该故障对应的入端 号;
然后根据该故障对应的入端口号来查询自身存储的本: ^业务转发表, 索引得到对应的出端口号、 入标签和出标签信息;
最后由上述得到的每对入标签和出标签分别确定的标答交换路径 LSP 作为与该故障相关的标签交换路径 LSPo
为了减小 MPLS系统中发现故障的节点的处理负荷,发现故障的节点 一般只向隧道标签(Tunnel Label )所确定的 LSP中下插传送故障信息, 所以上述所提及的入标签和出标签一般指隧道标签( Tunnel Label )。
同时上述提及的端口可以为逻辑端口 , 也可以为物理端口。
步骤 S20, 该发现故障的节点分别向步骤 S10所确定的每条相关的标 签交换路径 LSP中的宿端节点发送承载有故障信息的告警指示信号 AIS; 其中这里所述的告警指示信号 AIS可以为前向缺陷指^帧 FDI; 也可 以为包含有操作维护功能类型(即 OAM功能类型)、缺陷 型、故障节点 ID信息、 路径源端标识符(TTSI, Trail Termination Source Identifier )和 信号发送频率等信息的操作维护报文。
发现故障的节点可以根据相关的标签交换路径 LSP才应的出端口号 将告警指示信号 AIS由相应的出端口发出 (具体参照上述表格)。
步骤 S30, 每条相关 LSP的宿端节点在接收到告警指 ^信号 AIS后, 启动保护倒换处理。 为了接收到稳定的 AIS信号, 本实施例采用的是在接 收到相应数目的 AIS后, 启动保护倒换处理的操作, 即 ^据稳定接收的 AIS, 由发生故障的工作 LSP切换到保护 LSP上继续工作, 每条相关 LSP 上的宿端节点要接收相应数目的 AIS后才进行保护倒换处理,其目的是去 除抖动,待确定工作 LSP在确实出现故障的情况下,才进行保护倒换处理, 以避免在误接收 AIS的情况下, 就启动保护倒换处理, 浪铲系统的处理资 源。 在其它的实施例中, 保护倒换处理是在连续接收到相应数目的 AIS后 启动的。
为了在上述由 MPLS系统中的任意节点发现故障,并^故障信息迅速 向后扩展到相关的 LSP的宿端节点, 使宿端节点能够及时发现 LSP的故 障状态, 进行快速保护倒换处理的 J^ii上, 更进一步使保 倒换处理速度 能够达到现在电信级的保护倒换处理要求, 并使对系统业务带宽的影响降 至最小, 本发明又进而提出可以预先设置一个发送规则, 使发现故障的节 点以该设置的发送规则发送故障信息; 请参阅图 5, 该图是在本发明多协 议标签交换系统中的保护倒换方法中设置发送规则的实施例流程图; 其实 施过程主要为:
步骤 S110, 预先设置能够满足电信级保护倒换时间要求的发送规则; 步骤 S120, 由多协议标签交换 MPLS 系统中的任意节点发现故障, 并查找与故障相关的标签交换路径 LSP;
步骤 S130, 发现故障的节点分别以步骤 S110预先设置的发送规则向 每条相关的标签交换路径 LSP 中的宿端节点发送承载有故障信息的告警 指示信号 AIS;
步骤 S140, 每条相关 LSP的宿端节点在连续接收相应数目的告警指 示信号 AIS后, 启动保护倒换处理。
例如: 当步骤 S110中所述电信级保护倒换时间的要求是在 N毫秒内 完成保护倒换处理时(所述 N为正整数, 一般为 50 ), 并当步骤 S140中 所述宿端节点在连续接收 m个告警指示信号 AIS后, 启动保护倒换处理
(所述 m为自然数, 一般为 3 )的情况下, 则所设置的发送规则为: 故障 发生初始状态以大于等于^ ^个 /秒的发送速度连续发送 m个告警指示
N
信号后, 即在保证产生故障的 LSP的宿端节点能够及时接收到 m个告警 指示信号, 快速完成保护倒换处理后, 降低发送速率发送告警指示信号 AIS (如可以降至以 1个 /秒的发送速度发送告警指示信号 AIS )直至故障 恢复时终止。
这样依据上述的发送规则,既可以保证发生故障的 LSP的宿端节点肯 够及时接收到要求的 m个 AIS后进行保护倒换处理, 也可以尽量减小 系统业务带宽的占用。
请参阅图 6, 该图是在本发明多协议标签交换系统中的保护倒换方^ 中一个节点发现故障, 并相应查找与故障相关的 LSP的过程示意图; 图 6 中标签交换路由器 LSR/标签边缘路由器 LER表示处于一条 LSP上的任一 LSR或者 LER设备, 如果该设备的一个物理或者逻辑端口的物理层或者 数据链路层出现故障, 如以太网端口 link down, 或者多业务传送平台 MSTP中的虚级联组 VCG中的某个虚级联 VC失效, 或者通用成帧规程 GFP出现帧同步丟失告警等等 MPLS系统服务层中的信号失效(SF, Signal fail )状态, 该 LSR/LER设备利用本发明方案执行如下过程:
1、 该 LSR/LER发现自身的 Port m端口 link down;
2、 该 LSR/LER利用发生断开故障的 Port m端口的端口号 m查询自 身存储的本地业务转发表,索引并获得与断开端口 m相关的 LSP,分别为:
LSP Q, LSP M和 LSP N等三条路径;
3、 该 LSR/LER根据三条相关的 LSP的优先级顺序, 分别在 LSP Q, LSP M和 LSP N上按照一定的发送规则下插承载有与该端口 m对应的物 理链路失效的 AIS信息(其 AIS可以是一种新定义的 OAM帧, 即上述包 含有 OAM功能类型、缺陷类型、故障节点 ID信息、路径源端标识符 TTSI 和信号发送频率等信息的操作维护报文,也可以是标准中已有的 FDI帧), 其一定的发送规则是指: LSR/LER在发现故障的初期 ,快速发送 AIS信息, 如可以在故障初期每 5毫秒或者每 10ms发送一个 AIS信息; 并在连续发 送可以保证 LSP中的宿端节点能够正确识别 AIS信息后,如连续发送了 5 个 AIS信息或者 10个 AIS信息后, 其发送速度降低, 如可以降低至 1个 / 秒, 直到故障恢复状态为止。这样的发送规则即可以保证在 LSP出现故障 的情况下, 快速地将故障通告给该 LSP 的宿端节点, 还可以减小对 LSP 有效业务带宽的占用。
在 MPLS系统中, 发现故障的节点在获知故障消除后, 还可以进而向 相关的 LSP中的宿端节点发送故障消除信号,以使宿端节点在接收到故障 消除信号后, 能够启动相应的保换倒换恢复处理过程, 即从保护 LSP再回 到工作 LSP上工作。 请参阅图 7, 该图是本发明多协议标签交换系统中的 保护倒换方法发送故障消除信号的实施例流程图; 其实现过程如下:
步骤 S210, 由多协议标签交换 MPLS 系统中的任意节点发现故障, 并查找与故障相关的标签交换路径 LSP;
步骤 S220, 发现故障的节点分别以步骤 S110预先设置的发送规则向 每条相关的标签交换路径 LSP 中的宿端节点发送承载有故障信息的告警 指示信号 AIS;
步驟 S230, 每条相关 LSP的宿端节点在连续接收相应数目的告警指 示信号 AIS后, 启动保护倒换处理过程, 即从工作 LSP切换到保护 LSP 上继续工作;
步驟 S240,发现故障的节点在获知故障消除后,分别向每条相关的标 签交换路径 LSP中的宿端节点发送故障消除信号;
步骤 S250, 每条相关 LSP的宿端节点在接收到故障消除信号后, 启 动保护倒换恢复处理过程, 即从保护 LSP切换到工作 LSP上继续工作。
请参阅图 8, 该图是本发明多协议标签交换系统中的保护倒换方法在 MPLS系统拓朴图中的实施过程示意图; 假设图 8中的 LSP Q是由 LER 1 经过 LSR 2、 LSR 3到 LER 3的一个 LSP,当 LSR2和 LER 1之间的 MPLS 服务层出现故障时, LSR2根据该故障得到受影响的端口号, 并根据受影 响的端口号查询本地业务转发表, 索引到受该故障影响的各条 LSP, 假设 LSP Q是其中受该故障影响的一条 LSP; 继而 LSR 2会向出标签为 40, 出 端口为 LSR 2和 LSR 3之间的一个端口的 LSP Q路径中下插承载有相应故 障信息的 AIS信息, 如 FDI帧等; LER 3接收到相应数目的 AIS信息后, 确定相应 LSP出现故障, 启动相应 LSP的保护倒换处理流程。 这里同理 为了减少 MPLS系统中的中间节点的处理负荷,发现故障的节点只向由最 外层标签(即 Tunnel Label标签)组成的 LSP中下插 AIS信息。
以上所述仅是本发明的优选实施方式, 应当指出, 对于本技术领域的 普通技术人员来说, 在不脱离本发明原理的前提下, 还可以做出若干改进 和润饰, 这些改进和润饰也应视为本发明的保护范围。

Claims

权 利 要 求
1、 一种多协议标签交换系统中的保护倒换方法, 其特征在于, 包括步骤:
( 1 ) 由多协议标签交换系统中的任意节点发现故障, 并查找与故障相 关的标签交换路径;
( 2 ) 发现故障的节点分别向每条相关的标签交换路径中的宿端节点 发送承载有故障信息的告警指示信号;
( 3 ) 所述宿端节点在接收到告警指示信号后, 启动保护倒换处理。
2、 根据权利要求 1 所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述步骤( 1 )之前还包括步骤:
( P1 )预先设置能够满足电信级保护倒换时间要求的发送规则; 步骤(2 ) 中发现故障的节点以所述发送规则分别向每条相关的标签 交换路径中的宿端节点发送承载有故障信息的告警指示信号。
3、 根据权利要求 2所述的多协议标签交换系统中的保护倒换方法, 其特 征在于,
步骤(P1 )中所述电信级保护倒换时间要求是在 N毫秒内完成保护倒 换处理, 所述 N为正整数;
所述发送规则为初始以大于等于^ ^个 /秒的发送速度连续发送 m
N
个告警指示信号后, 降低发送速度发送告警指示信号至故障恢复时终止; 步驟(3 ) 中所述宿端节点在连续接收 m个告警指示信号后, 启动保 护倒换处理, 所述 m为自然数。
4、 根据权利要求 1 所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述发现故障的情况具体包括:
所述节点发现自身物理端口对应的物理链路失效; 或
发现自身多业务传送平台中的虚级联组失效; 或
发现自身通用成帧规程出现帧同步丢失告警。
5、 根据权利要求 1 所述的多协议标签交换系统中的保护倒换方法 , 其特 征在于 , 所述查找与故障相关的标签交换路径的过程具体包括: ( 21 )建立能够才 据入端口号查找到出端口号、 入标签和出标签的映射 关系表;
( 22 )确定所述故障对应的入端口号;
( 23 )根据故障对应的入端口号查询所述映射关系表, 得到对应的出端 口号、 入标签和出标签信息;
( 24 )由得到的每对入标签和出标签分别确定的标签交换路径作为与故 障相关的标签交换路径。
6、 根据权利要求 1 所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述查找与故障相关的标签交换路径的过程具体包括:
( 2a )确定所述故障对应的入端口号;
( 2b )根据故障对应的入端口号查询本地业务转发表, 得到对应的出端 口号、 入标签和出标签信息;
( 2c ) 由得到的每对入标签和出标签分别确定的标签交换路径作为与故 障相关的标签交换路径。
7、 根据权利要求 5或 6所述的多协议标签交换系统中的保护倒换方法, 其特征在于, 所述入标签和出标签为隧道标签。
8、 根据权利要求 5或 6所述的多协议标签交换系统中的保护倒换方法, 其特征在于, 所述步骤(2 ) 中, 发现故障的节点 居相关的标签交换路 径对应的出端口号将告警指示信号由相应的出端口发出。
9、 根据权利要求 8所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述端口为逻辑端口或物理端口。
10、 根据权利要求 1所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 还包括步骤:
(.4 )发现故障的节点在故障消除后, 分别向每条相关的标签交换路径 中的宿端节点发送故障消除信号;
( 5 )所述宿端节点在接收到故障消除信号后, 启动保护倒换恢复处理。
11、 根据权利要求 1 ~ 6任意权利要求所述的多协议标签交换系统中的保 护倒换方法, 其特征在于, 所述告警指示信号为:
前向缺陷指示帧; 或 包含有操作维护功能类型、 缺陷类型、 故障节点 ID信息的操作维护报文。
12、 根据权利要求 1所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述宿端节点在接收多个告警指示信号后, 启动保护倒换处理。
13、 根据权利要求 1所述的多协议标签交换系统中的保护倒换方法, 其特 征在于, 所述宿端节点在连续接收多个告警指示信号后, 启动保护倒换处 理。
PCT/CN2005/001532 2004-09-30 2005-09-22 Procede de rearrangement de protection dans un systeme mpls WO2006034629A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CNB2004100793550A CN100397826C (zh) 2004-09-30 2004-09-30 多协议标签交换系统中的保护倒换方法
CN200410079355.0 2004-09-30

Publications (1)

Publication Number Publication Date
WO2006034629A1 true WO2006034629A1 (fr) 2006-04-06

Family

ID=36118570

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2005/001532 WO2006034629A1 (fr) 2004-09-30 2005-09-22 Procede de rearrangement de protection dans un systeme mpls

Country Status (2)

Country Link
CN (1) CN100397826C (zh)
WO (1) WO2006034629A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102315972A (zh) * 2011-10-14 2012-01-11 杭州华三通信技术有限公司 用于实现lsp倒换的方法和装置

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101227370B (zh) * 2008-02-05 2011-01-19 中兴通讯股份有限公司 传送多协议标签交换共享保护环故障的处理方法
CN101540637B (zh) * 2008-03-21 2013-01-09 北京润光泰力科技发展有限公司 网络故障的定位方法和系统
CN101997747B (zh) * 2009-08-21 2014-12-10 中兴通讯股份有限公司 故障lsp信息通告的方法和装置
CN101860461B (zh) * 2010-05-17 2015-10-21 中兴通讯股份有限公司 一种在汇聚链路中进行用户端口管理的方法及系统
CN101986620A (zh) * 2010-11-01 2011-03-16 中兴通讯股份有限公司 源路由环网保护的快速倒换方法、装置及系统
CN102006222B (zh) * 2010-11-16 2015-06-24 中兴通讯股份有限公司 一种业务链路切换方法及装置
CN102142982A (zh) * 2010-11-23 2011-08-03 华为数字技术有限公司 一种转发平面故障修复方法和装置
CN102299864B (zh) * 2011-10-08 2014-11-26 杭州华三通信技术有限公司 一种应用于二层虚拟专用网络的报文转发的方法和装置
CN102377601B (zh) * 2011-10-14 2014-03-26 杭州华三通信技术有限公司 一种lsp故障通告方法和装置
DE112011105843T5 (de) 2011-11-14 2014-08-28 Intel Corporation Pfaddiversität in einem verbindungsorientierten Netzwerk
CN102801558B (zh) * 2012-07-30 2016-03-30 中国联合网络通信集团有限公司 基于分组传送网的链路保护方法、节点和系统
CN103916258A (zh) * 2013-01-07 2014-07-09 中兴通讯股份有限公司 告警指示信号告警清除方法、服务层及客户层维护实体组端点
CN103281244B (zh) * 2013-06-17 2016-12-28 瑞斯康达科技发展股份有限公司 一种多协议标签交换网络的业务承载方法及装置
CN103905274A (zh) * 2014-03-25 2014-07-02 华为技术有限公司 一种故障信息的传递方法和设备
CN105099903B (zh) * 2014-04-15 2018-12-07 华为技术有限公司 光包交换系统的链路确认方法、装置及系统
CN103997422B (zh) * 2014-05-05 2019-06-28 大唐移动通信设备有限公司 一种ip接口板的故障处理方法和装置
CN107995008B (zh) * 2016-10-27 2021-07-06 中兴通讯股份有限公司 一种业务告警处理方法、装置及系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020060985A1 (en) * 2000-11-22 2002-05-23 Lee Jae Yong Method for high speed rerouting in multi protocol label switching network
US20030112749A1 (en) * 2001-12-18 2003-06-19 Brian Hassink Methods, systems, and computer program products for detecting and/or correcting faults in a multiprotocol label switching network by using redundant paths between nodes
CN1466832A (zh) * 2002-01-02 2004-01-07 ƽ 用于快速重路由的隐式共享带宽保护

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1145333C (zh) * 2001-06-27 2004-04-07 华为技术有限公司 多协议标签交换快速保护倒换方法
JP2003060680A (ja) * 2001-08-14 2003-02-28 Nippon Telegr & Teleph Corp <Ntt> ネットワーク障害処理システムと方法およびノードならびにプログラムと記録媒体

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020060985A1 (en) * 2000-11-22 2002-05-23 Lee Jae Yong Method for high speed rerouting in multi protocol label switching network
US20030112749A1 (en) * 2001-12-18 2003-06-19 Brian Hassink Methods, systems, and computer program products for detecting and/or correcting faults in a multiprotocol label switching network by using redundant paths between nodes
CN1466832A (zh) * 2002-01-02 2004-01-07 ƽ 用于快速重路由的隐式共享带宽保护

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102315972A (zh) * 2011-10-14 2012-01-11 杭州华三通信技术有限公司 用于实现lsp倒换的方法和装置
CN102315972B (zh) * 2011-10-14 2013-12-25 杭州华三通信技术有限公司 用于实现lsp倒换的方法和装置

Also Published As

Publication number Publication date
CN1756182A (zh) 2006-04-05
CN100397826C (zh) 2008-06-25

Similar Documents

Publication Publication Date Title
WO2006034629A1 (fr) Procede de rearrangement de protection dans un systeme mpls
US7881184B2 (en) Reverse notification tree for data networks
US8130637B2 (en) Method and apparatus for detecting MPLS network failures
Huang et al. Building reliable MPLS networks using a path protection mechanism
US8737203B2 (en) Method for establishing an MPLS data network protection pathway
US7742400B2 (en) Method and system for detecting link failure between nodes in a hybrid network
EP0961518B1 (en) Operator directed routing of connections in a digital communications network
EP2028774B1 (en) Method, system and node device for realizing service protection in the automatically switched optical network
US7821971B2 (en) Protection providing method and customer edge apparatus
CN101621497B (zh) 一种多层网络中的业务保护的方法和系统
US20130272114A1 (en) Pseudo wire switching method and device
WO2006039865A1 (fr) Procede permettant de mettre en oeuvre une decouverte automatique de structure de topologie dans le reseau boucle mpls
EP2501084B1 (en) Transmission multi-protocol label switching network system and link protection method
WO2015184868A1 (zh) 一种服务层信号失效检测装置及方法
EP2239956B1 (en) Method, apparatus and system for ip/optical convergence
WO2011020257A1 (zh) 故障lsp信息通告的方法和装置
US7848246B2 (en) Method and system for confirming connection of layer-1 label switched path(L1-LSP) in GMPLS-based network
WO2009067909A1 (fr) Procédé, dispositif de nœud de réseau et système pour configurer automatiquement une table de réglage silencieux
EP2117199B1 (en) Transmission method, system and router based on the border gateway protocol
WO2004114594A1 (fr) Procede de reprise en secours du service d&#39;un dispositif dans un reseau de groupe atm (mode de transfert asynchrone)
WO2005050933A1 (en) Point-to-point route monitoring in a packet-based core network
WO2011020261A1 (zh) 一种环网双发机制的保护方法和系统

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase