CN103891215B - Fault detect in the end-to-end protection solution of multiprotocol label switching multicast label switched path - Google Patents

Fault detect in the end-to-end protection solution of multiprotocol label switching multicast label switched path Download PDF

Info

Publication number
CN103891215B
CN103891215B CN201280049788.4A CN201280049788A CN103891215B CN 103891215 B CN103891215 B CN 103891215B CN 201280049788 A CN201280049788 A CN 201280049788A CN 103891215 B CN103891215 B CN 103891215B
Authority
CN
China
Prior art keywords
lsp
node
fault detection
datum plane
notification message
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201280049788.4A
Other languages
Chinese (zh)
Other versions
CN103891215A (en
Inventor
赵强林
陈莹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority claimed from PCT/US2012/059800 external-priority patent/WO2013055950A1/en
Publication of CN103891215A publication Critical patent/CN103891215A/en
Application granted granted Critical
Publication of CN103891215B publication Critical patent/CN103891215B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

On the one hand, the present invention includes, in the root node on label switched path, a kind of computer program, including the computer executable instructions being stored in non-transitory medium, when processor performs described instruction, described root node performs following operation: create the fault detection conversation based on the first datum plane occurring have unactivated state on the first label switched path (LSP) together with at least one leaf node, receive the notification message of the predetermined quantity from described leaf node, the notification message instruction of wherein said predetermined quantity, when second processor is sent to leaf node, the fault of generation fault detection conversation based on the second datum plane on the 2nd LSP;And after the notification message receiving described predetermined quantity, the state that will occur in described based on the first datum plane the fault detection conversation on a described LSP is changed into inactive.

Description

Fault detect in the end-to-end protection solution of multiprotocol label switching multicast label switched path
CROSS REFERENCE TO RELATED application
Invention entitled " the multiprotocol label switching multicast mark that application claims is submitted on October 11st, 2011 Sign switching path end-to-end protection solution in fault detect " No. 61/545897 U.S. interim The earlier application priority of application case, the mode that the content of these earlier applications introduces in full is expressly incorporated herein In.
Background technology
Multiprotocol label switching (mpls) tag distribution protocol (LDP) can be used in a network Create point-to-multipoint (P2MP) and how point-to-multipoint (MP2MP) label switched path.For creating The LDP superset building P2MP or MP2MP LSP can be described as multiple spot LDP (mLDP), and it is mutually Entitled " the point-to-multipoint and mark of multipoint-to-multipoint label switch paths of networking project task groups (IETF) Sign distribution protocol extension " Request for Comment (RFC) 6388 in illustrate, its content is with the side quoted Formula is expressly incorporated herein in this.Some upper tag distribution (ULA) technology " is used for IETF is entitled LDP MPLS upper tag distribution " RFC6389 in illustrate, its content is to quote Mode be expressly incorporated herein this.
Service provider uses mLDP to continue in an mpls network to dispose and multicast application in real time.Obviously, Need these real-time application are protected, and make when breaking down switching time the shortest.Protect at present The way protecting business and high application includes precomputation and creates backup path.FAXIA is done this, one Denier detects fault on main path, it should use control plane that flow is rerouted to backup path On.But, when nodes break down on a P2MP LSP, the second external network or client End be used for determining fault and be switched to receive the delay of the second Egress node of flow may be longer.? In some systems, this delay is probably unacceptable, such as, for such as Internet protocol (IP) electricity For real time services such as (IPTV).
Summary of the invention
On the one hand, the present invention includes a kind of computer in the root node on label switched path Program product, including the computer executable instructions being stored in non-transitory medium, when processor is held During the described instruction of row, described root node performs following operation: create together with at least one leaf node Occur have on the first label switched path (LSP) unactivated state based on the first datum plane Fault detection conversation;Receive the notification message of the predetermined quantity from described leaf node, Qi Zhongsuo State the notification message instruction of predetermined quantity, when the second processor is sent to leaf node, occur at the 2nd LSP On the fault of fault detection conversation based on the second datum plane;And receiving described predetermined number After the notification message of amount, will occur in described based on the first datum plane the event on a described LSP The state of barrier detection session changes into activation.
On the other hand, the present invention includes, at the leaf of the network comprising multiple label switched path In child node, a kind of networking component includes processor, exists for creating generation together with the first head node Fault detection conversation based on the first datum plane on first label switched path (LSP), wherein Described fault detection conversation based on the first datum plane has state of activation, and is used for and the second cephalomere Point creates the fault detection conversation based on the second datum plane occurred on the 2nd LSP together, wherein Described fault detection conversation based on the second datum plane has unactivated state;And in trigger event Transmitting notification messages to described second head node after generation, the instruction of wherein said notification message occurs Transmission fault on oneth LSP.
It yet still another aspect, the present invention includes a kind of including multiple mark with at least one leaf node Signing the method in the network system of switching path, described method is for breaking down after in a LSP Switch the transmission of LSP to the 2nd LSP, including: create a LSP, on a described LSP The first head node and at least one leaf node between create fault based on the first datum plane inspection Surveying session, wherein said fault detection conversation message based on the datum plane described LSP of instruction is Activate;Create the 2nd LSP;Receive the instruction indicating a described LSP to break down, Qi Zhongsuo State instruction to include not receiving at least one fault based on datum plane from described first head node Detection conversation message;And be sent to indicate a described LSP from described leaf node by notification message The second head node broken down.
In conjunction with drawings and claims, can be more clearly understood that from the following detailed description these and Other features.
Accompanying drawing explanation
In order to be more fully understood from the present invention, with reference now to the letter carried out below in conjunction with accompanying drawing and detailed description Describing, wherein same reference numerals represents same section.
Fig. 1 describes the schematic diagram of an embodiment of a kind of label switching system;
Fig. 2 describe MPLS multicast LSP(mLSP) end-to-end protection solution in fault inspection The flow chart of one embodiment of survey method;
Fig. 3 describes an embodiment of the example network before the head node on main LSP breaks down;
Fig. 4 describes an embodiment of the example network when host node on main LSP breaks down;
Fig. 5 describes an embodiment of the example network after the host node on main LSP breaks down;
Fig. 6 describes a kind of be applicable to one or more embodiments implemented assembly disclosed in this invention Typical general-purpose network component.
Detailed description of the invention
Although first it should be understood that the exemplary embodiment party having provided below one or more embodiments Formula, system disclosed by the invention and/or method can be by other technology implementations that are known or that exist multiple. The present invention should in no way be limited to described illustrative embodiment described below, graphic and technology, comprises this Literary composition is illustrated and the exemplary design that describes and embodiment, but can be at the model of appended claims Enclose and amendment in the full breadth of its equipollent.
It is currently used for the slow again based on the signaling mechanism complexity controlling plane of fault detect, carries out fault inspection Survey and generally about need one second.The present invention introduces one between head node and the nodes at ends of multicast LSP Plant the most simple and quick signaling mechanism based on datum plane.The work of present invention mechanism is similar to heart beating, Such as, from head node to leaf node, for main LSP and the pulse detection information from LSP.Work as master When LSP heart beating occurs to interrupt, this system can send a notice to back up LSP tree head, can be by backup LSP starts to back up flow and forwards.Once nodes at ends receives the backup flow on backup LSP, and tail end saves Point can notify the head node of the front main LSP broken down, and this switching has occurred and that.When breaking down it The head node of front main LSP receives switching notice, stops the flow along main LSP before and forwards, Thus complete switching.Signaling mechanism based on datum plane also can indicate that the leaf broken down on main tree Number.
Fig. 1 describes an embodiment of label switching system 100, at least a part of which can some assemblies it Between create a plurality of P2P LSP and P2MP LSP.Described P2P LSP and P2MP LSP can be used for transmitting Data traffic, as used bag and bag label to route.Label switching system 100 can include Tag switching Network 101, this network can be a kind of use bag or frame carry out data transmission along network path or route Packet network.This bag can route along described path or exchange, and can create such as MPLS or logical With MPLS(GMPLS) etc. label switching protocol.
Label exchange network 101 can include that multiple fringe node, multiple fringe nodes include the first entrance joint Point the 111, second Ingress node 112, multiple first Egress node 121 and multiple second Egress node 122.When the P2MP LSP in label exchange network 101 includes an inlet and an outlet fringe node, first Ingress node 111 and the second Ingress node 112 can be described as root node or head node, the first Egress node 121 Leaf node or nodes at ends is can be described as with the second outlet junctions 122.Additionally, label exchange network 101 Can include multiple internal node 130, this internal node can communicate with each other, it is possible to fringe node Communicate.It addition, the first entrance contact 111 and the second Ingress node 112 can with such as Internet protocol (IP) source node 145 in first external network such as network 140 communicates, and external network is alternatively coupled to Label exchange network 101.Additionally, the first Egress node 121 and the second Egress node 122 can be with purposes Ground node 150 or other networks 160 communicate.Consequently, it is possible to the first Ingress node 111 and second Data such as packet, can be transferred to destination's node 150 from external network 140 by Ingress node 112.
In one embodiment, fringe node and internal node 130(are referred to as network node) can be Support to be carried out any equipment or the assembly of bag transmission by label exchange network 101.Such as, described network Node includes switch, router, or the multiple combination of this kind equipment.Each network node may each comprise For receiving the receptor of bag from other network nodes, in order to determine these bags are sent to which network joint The processor of point or other logic circuits, and these bags are transferred to the emitter of other network nodes. In certain embodiments, at least some network node can be that LSR, this LSR can be used for revising or updating The label of the bag of transmission in label exchange network 101.Further, at least some fringe node can be mark Signing edge router (LER), described tag edge router can be used for inserting or removing at label exchange network The label of the bag of transmission between 101 and external network 140.
Label exchange network 101 can include a P2MP LSP105, can create a P2MP LSP105, to be multicasted to destination's node 150 or other nets by data traffic from the first external network 140 Network 160.Oneth P2MP LSP105 can include that the first Ingress node 111 and at least some first export joint Point 121.Oneth P2MP LSP105 is illustrated by solid arrow line in FIG.Generally, for protection first P2MP LSP105, in case it occurs that link or nodes break down, label exchange network 101 can include 2nd P2MP LSP106, P2MP LSP106 can include the second Ingress node 112 and at least some second Egress node 122.2nd P2MP LSP106 is illustrated by dotted arrow line in FIG.Each second goes out Mouth node 122 can be with the first Egress node 121 groups pairs in a described P2MP LSP105.Second P2MP LSP106 also includes some identical or diverse internal nodes 130.Described 2nd P2MP LSP106 can provide one article of backup path leading to a P2MP LSP105, can be used at P2MP When the networking component of LSP105 breaks down, flow is forwarded to first from the first external network 140 P2MP LSP105 or the 2nd P2MP LSP106, as being forwarded to Egress node 123.
When the component failure of P2MP LSP105, via corresponding the 2nd P2MP LSP106 couple Flow re-routes may cause flow propagation delay.Even when the stream entrained by the 2nd P2MP LSP106 Measure identical with a P2MP LSP105, when the networking component of a P2MP LSP105 breaks down, Oneth P2MP LSP105 or the 2nd P2MP LSP106 is used for determining this fault and being switched to transmission stream The delay of the backup path of amount may be longer.This delay is probably unacceptable in some systems, example As, for the real time services such as such as IPTV.
Fig. 2 describes of the fault detection method in the end-to-end protection solution of MPLS mLSP The flow chart of item embodiment.Process 200 can be advocated peace from the beginning of LSP by establishment from 201.Such as, figure The first Ingress node 111 in 1 can be created between the first Ingress node and 111 and Egress node 123 Building a P2MP LSP105, the second Ingress node 112 can be at the second Ingress node 112 and outlet joint The 2nd P2MP LSP106 is created between point 123.
As shown at 203, fault detection conversation based on datum plane can be created on LSP1 and LSP2, Also referred herein as heart beating.Fault detection conversation based on datum plane may be included in predetermined time interval and passes Defeated multiple message be used for (a) notice downstream node any bar LSP be activate and/or be main LSP and (b) checking path seriality.In one embodiment, the event based on datum plane of LSP1 and LSP2 Barrier detection session can include that two kinds individually make control message label be set to activate and nonactivated unidirectional Fault detect (UFD) session.UFD is for detecting two-way turn of MPLS LSP data plane failure Send out the subset of detection (BFD) agreement, identical agreement generally can be utilized as traditional BFD.If Meter BFD detects Egress node (example for Egress node (such as, first ingress 111 of Fig. 1) Such as, the Egress node 123 of Fig. 1) disconnection that connects, and provide entrances to save by some optional mechanism Point follows the trail of connectivity.Under UFD agreement, Ingress node need not or not receive response message.BFD Being known in the prior art with UFD agreement, its additional information is from entitled " the two-way converting inspection of IETF Survey (BFD) " RFC5880 in obtain, during its content is hereby incorporated herein by this.
LSP1 and LSP2 can be created the most in a certain order.Once create relevant LSP, Creating LSP heart beating, it is the UFD in Fig. 2 embodiment the most in any order Session.Such as, in another embodiment, until detection LSP1 just creates the heart of LSP2 after breaking down Jump.When the heart beating of LSP2 creates, and one or more leaf nodes can receive from LSP1 and LSP2 Heart beating.Such as, in FIG, Egress node 123 can receive from a LSP105 The heart beating of one Ingress node 111, it is also possible to receive from the second Ingress node 112 on the 2nd LSP106 Heart beating.
By using fault detection conversation agreement based on datum plane, the leaf node in system can be used for Expect to receive the periodic heartbeat of the root node from LSP.Leaf node can save from root by not receiving The expection heart beating of the predetermined quantity of point detects fault.
As shown in 205, after fault detection, the Flow Fault on LSP1 can be notified by leaf node It is sent to the root node of LSP2.Such as, in FIG, if Egress node 123 is at a LSP105 On the heartbeat interrupt from the first Ingress node 111 detected, Flow Fault can be led to by Egress node 123 Know and be sent to the second Ingress node 112.As shown in 207, after LSP1 receives Flow Fault notice, The root node of LSP2 can start to be forwarded to the flow on LSP2 leaf node the state by relevant heart beating It is set to activate.Such as, in FIG, if the second Ingress node 112 receives LSP105 from Egress node 123 On Flow Fault notice, the second Ingress node 112 can be at the second Ingress node 112 and Egress node The UFD label of oneself is set to by the UFD session between 123 activate.As shown in 209, work as leaf Nodal test to the heart beat status on LSP change time, leaf node can be ignored from LSP1 Flow and can notify that the root node of LSP1 there occurs that LSP1 fault and flow are forwarded on LSP2. Such as, in FIG, if Egress node 123 detect the UFD label of the second Ingress node 112 from Inactive becoming is activated, and switching notice can be sent to the first Ingress node 111 by Egress node 123.As Shown in 211, after receiving this notice, its heart beat status can be set to inactive by the root node of LSP1 And stop forwarding the traffic to leaf node, thus complete switching.Such as, in FIG, if first Ingress node 111 receives switching notice from Egress node 123, and the first Ingress node 111 can be by certainly Oneself UFD label is set to inactive for being transmitted on LSP105.
Before Fig. 3,4 and 5 describe the head node fault on main LSP, time and example network afterwards A kind of embodiment.Network shown in Fig. 3-5 can be similar with the network shown in Fig. 1.Fig. 3's Tree starts from external transmission source 300, and such as, IPTV inputs, and it may be used for by Ingress node 302 Transmission data.Ingress node 302 can divide data into two paths.Main path, LSP1340 can With from root 304(PE1) (also referred herein as head node, root node or tree head) from the beginning of, can pass through Internal node 306(P1) and 308(P2) transfer data to leaf node 310(PE2, PE3, PE4 and PE5), the most also become nodes at ends.From path, LSP2350, can be from root 314(PE1 ') Start, it is possible to for by node 316(P3) and 318(P4) transfer data to leaf node 310(PE2, PE3, PE4 and PE5).Leaf node 310(PE2, PE3, PE4 and PE5) can For use as Egress node, such as, transferring data to can be as the external transmission purpose of IPTV output Ground 320.As it can be seen, the network node being included in LSP1340 and LSP2350 can different and The most non-intersect.Similarly, LSP1340 and LSP2350, but the while of being not both, it will usually Data are transmitted to outside transmission destination 320.In figs. 3-5, LSP1340 is main LSP, LSP2350 is backup LSP.Fig. 4 describes the embodiment of Fig. 3 that root 304 breaks down.Fig. 5 retouches State the embodiment of the Fig. 4 being switched to LSP2350.In Figure 5, root 314(PE1 ') it is following institute The head node of the new main LSP2350 stated.
With reference to Fig. 3-5, the operation of system may begin at tree 304 and 314 at head node 304,314 And between leaf node 310, create LSP and fault detection conversation based on datum plane, also referred to as heart beating. In the embodiment of Fig. 3-5, heart beating is UFD session, but other fault detects based on datum plane Session protocol is to allow.On LSP1340 and LSP2350, once transmit heart beating, can be simultaneously by head Heart beating is continually transmitted on leaf node 310 by node 304 and 314, and leaf node 310 likely connects The continuous heart beating receiving head node 304 and 314 transmission.Root 304(PE1) heart can be transmitted by state of activation Jump, such as, can save to leaf by sending UFD message together with being provided with activation label on LSP1340 Point 310.Root 314(PE1 ') can by unactivated state send heart beating, such as can by with LSP2350 On be provided with inactive label and send UFD message together to leaf node 310.Square 201 He in figure 203 generally describe these operations.
If internal node 306(P1) break down, as described in Figure 4, leaf node 310 can Can not receive from tree head 304(PE1) UFD message.When leaf node 310 is in the UFD session phase Between do not receive the UFD message of predetermined quantity, leaf node 310 can classify as path by not receiving therefore Notification message also is sent to set head by barrier, and its UFD message is still received by leaf node 310.In Fig. 2 Square frame 205 generally described this operation.The UFD message not received needed for trigger notice message Quantity, can select according to required system sensitivity, and this also determines with demand simultaneously. Sensory system there may be the risk sending error notification.Less sensitive system may causing trouble response The risk of time lag.In one embodiment, the single UFD not received at single leaf node 310 Message can be with trigger notice message;In another embodiment, it may be necessary to 2-10 the UFD not received Message.Other embodiments may need not the receiving at the leaf node 310 of more than 10 UFD message triggers and sends a notification message.
As root 314(PE1 ') receive the notification message of predetermined quantity, root 314(PE1 ' from leaf node 310) Converting flow can be started.The quantity triggering the notification message needed for flow forwards can be according to required system Sensitivity selects, and this also determines with demand simultaneously.Sensory system there may be in error notification The risk that flow forwards is started in message.Less sensitive system may bring extraneous response time lag Risk.In one embodiment, single notification message can trigger forwarding, and in another embodiment, 2-10 notification message may be needed.Other embodiments need more than 10 of leaf node 310 Notification message triggers flow and forwards.Other embodiments may need from two or more leaf nodes At least one notification message of 310.Root 314(PE1 ') can have by sending on LSP2350 The UFD message activating label starts to forward to leaf node 310.Square frame 207 in Fig. 2 is generally These operations are described.Once root 314(PE1 ') start to forward and leaf node 310 on LSP2350 Detect that LSP2350 heart beat status becomes activating from un-activation, still receive flow from LSP1340 Arbitrary leaf node 310 can abandon from the bag of LSP1340 and utilize the bag from LSP2350.
The state that once leaf node 310 receives on LSP2350 becomes the heart for activating from un-activation Jumping, leaf node 310 can transmit notification messages to the head node of previous activation, i.e. switches Root 304(PE1).Square frame 209 in Fig. 2 generally described following operation: detection did not previously swash The heart beat status on LSP of living changes and switching notice is sent to the root node of previous activation LSP.One Denier is triggered, and such as leaf node 310 sends the notice of switching message, root 304(PE1) will stop Forward the flow on LSP1340, and the state of LSP1340 heart beating is changed into un-activation.In Fig. 2 Square frame 211 generally described these operation.In one embodiment, by root 304(PE1) receive Single handoff notification message can trigger and stop forwarding, and in another embodiment, it may be necessary to 2-10 Individual handoff notification message.Other embodiments need more than 10 notification messages from leaf node 310 Stop flow forwarding.Other embodiments may need at least one from two or more leaf nodes The notification message of 310.
Because this solution utilizes datum plane rather than controls plane, so fault correction time is permissible Calculate with millisecond rather than second.Fault correction time under the present invention is likely to be dependent on some questions, including But it is not limited to, the cycle of UFD message, triggers leaf node and give notice the required UFD not received The number of cycles of message, the delivery flow rate triggering backup LSP head node is changed required leaf node The quantity of notice, it is provided that notice triggers the delivery flow rate of backup LSP head node and is changed required joint Point quantity, receive have renewal state tag backup LSP head node send UFD message time Between, trigger the number switching this type of UFD message needed for message informing, and at main LSP head node Place receives switching message and completes to switch the required time.Fault correction time can be according to the needs of application Arbitrarily adjust.In the case of the faster fault correction time of needs, fault correction time can be less than about 200 milliseconds, less than about 100 milliseconds or less than about 10 milliseconds.Requiring the preferential of fault detection accuracy In the case of level is higher, longer fault correction time can be allowed to obtain multiple indicating fault.? In the case of requiring that fault correction time height is preferential, false error message and frequently switching can be accepted.
Above-mentioned networking component can be implemented, such as described in Fig. 1 and 3-5 in any general-purpose network component Those networking components, it has enough disposal ability, storage resource and network throughput capability to process it On necessary workload.Fig. 6 illustrates typical general-purpose network component 1000, and it is applicable to implement herein One or more embodiments of this disclosed assembly.It is permissible that networking component 1000 comprises processor 1002( It is referred to as central processor unit or CPU), described processor and the memory device communication comprising following item: Auxilary unit 1004, read only memory (ROM) 1006, random-access memory (ram) 1008, input/output (I/O) equipment 1010, and network access device 1012.Processor 1002 Can implement as one or more cpu chips, or can be one or more special IC And/or a part for digital signal processor (DSP) (ASIC).
Auxilary unit 1004 is generally by one or more disc drivers or erasable programmable ROM (EPROM) composition, and for the non-volatile memories of data.Additional storage 1004 may be used for Storage program, when selecting to perform these programs, described program will be loaded in RAM1008. The instruction read the term of execution that ROM1006 being for being stored in program and the data that may read. ROM1006 is non-volatile memory device, and its memory capacity is bigger relative to additional storage 1004 For memory capacity generally the least.RAM1008 is used for storing volatile data, and is possibly used for storage Instruction.The access speed of ROM1006 and RAM1008 generally depositing than additional storage 1004 Take speed fast.
The present invention discloses at least one embodiment, and those of ordinary skill in the field are to described embodiment And/or the feature of described embodiment make change, combine and/or revise all in scope disclosed by the invention In.The alternate embodiment obtained because combining, merge and/or omit the feature of described embodiment is also at this In the range of invention.It should be appreciated that the present invention the most clearly illustrates numerical range or restriction, this type of Clear and definite scope or restriction should include containing in above-mentioned scope or limiting (such as the model from about 1 to about 10 Enclose and include 2,3,4 etc.;Scope more than 0.10 includes 0.11,0.12,0.13 etc.) in similar number The iteration ranges of magnitude or restriction.Such as, whenever the open numerical range with lower limit Rl and upper limit Ru Time, any numeral in the range of specifically disclosing described in falling into.Specifically, specifically disclosed described scope In following numeral: R=Rl+k* (Ru-Rl), wherein k is with the change of 1% incremental increase from 1% to 100% Amount, i.e. k is 1%, 2%, 3%, 4%, 5% ..., 50%, 51%, 52% ..., 95%, 96%, 97%, 98%, 99% or 100%.Additionally, the most hereby disclose, two R defined above The defined any numerical range of value.Unless otherwise specified, term " about " represent numerical value thereafter ± 10% Scope.Relative to a certain key element of claim, the use that term " may select " represents that this key element is permissible Being " needs ", or can also be " unwanted ", both be in the range of described claim. Use such as include, comprise and have etc. wider term should be understood to provide to such as by " ... form ", The support of narrower term such as " substantially by ... composition " and " generally by ... composition ".Therefore, Protection domain is not by limitations described above, but is defined by the appended claims, and described scope comprises All equivalents of the subject matter of appended claims.Each and every claim are as the most public The content opened is incorporated in description, and claims are embodiments of the invention.In described disclosure The discussion of reference be not an admission that it is prior art, especially there is the earlier application in subject application Any reference of the publication date after priority date.All patents cited in the present invention, patent The disclosure of application case and publication is hereby incorporated herein by this hereby, and it provides and supplements this Exemplary, procedural or other details of invention.
Although the present invention has been provided that some embodiments, it should be appreciated that without departing from the present invention spirit or In the case of scope, system and method disclosed in this invention can embody with other particular forms many. The example of the present invention should be considered illustrative and not restrictive, and the present invention be not limited to Ben Wenben to The details gone out.Such as, various elements or assembly can combine in another system or merge, or some Feature can be omitted or do not implement.
Additionally, without departing from the scope of the invention, described in various embodiments and explanation for from Dissipate or individually technology, system, subsystem and method can be with other system, module, techniques or methods It is combined or merges.Show or be discussed as being coupled to each other or directly couple or communicate other can also Use electrically, mechanical system or other modes be by a certain interface, device or intermediate module coupling indirectly Connect or communicate.Other changes, substitute and examples of changing can by those skilled in the art without departing from Determine in the case of the most spiritual and disclosed scope.

Claims (20)

1. the fault detection method in the root node in the second label switching path LSP, the root node on described 2nd LSP performs to operate as follows:
The fault detection conversation based on the second datum plane occurring that there is on the 2nd LSP unactivated state is created together with at least one leaf node;
Receive the notification message of the predetermined quantity from described leaf node, the notification message instruction of wherein said predetermined quantity, when root node on oneth LSP is sent to described leaf node, the fault of generation fault detection conversation based on the first datum plane on a LSP;And
After the notification message receiving described predetermined quantity, the state that will occur in described based on the second datum plane the fault detection conversation on described 2nd LSP changes into activation.
Method the most according to claim 1, it is characterised in that described fault detection conversation based on the second datum plane is that unidirectional fault detects UFD session.
Method the most according to claim 1, it is characterised in that the notification message of described predetermined quantity is 1 to 10.
Method the most according to claim 1, it is characterised in that the time between the first notification message and change state in the notification message of described reception predetermined quantity is less than 200 milliseconds.
5., in comprising the leaf node of network of multiple label switched path, a kind of networking component includes:
A kind of processor, is used for:
Creating the fault detection conversation based on the first datum plane occurred in the first label switching path LSP together with the first head node, wherein said fault detection conversation based on the first datum plane indicates the state of activation of a described LSP;
Creating fault detection conversation based on the second datum plane together with the second head node on the 2nd LSP, wherein said fault detection conversation based on the second datum plane indicates the unactivated state of described 2nd LSP;And
Transmitting notification messages to described second head node after trigger event occurs, there is the transmission fault on a LSP in the instruction of wherein said notification message.
Networking component the most according to claim 5, it is characterised in that the state that described processor is further used for receiving in the fault detection conversation based on the second datum plane that the state of the 2nd LSP is updated to activate updates.
Networking component the most according to claim 6, it is characterised in that described processor is further used for the first head node that switching state message is sent to indicate described 2nd LSP state to change.
Networking component the most according to claim 7, it is characterised in that described processor is further used for reception and the state of a described LSP is updated to nonactivated switching renewal.
Networking component the most according to claim 6, it is characterised in that described processor is further used for ignoring the flow of a described LSP after receiving state and updating.
Networking component the most according to claim 6, it is characterised in that described processor is further used for sending additional informing message after trigger event at a predetermined interval until the state that receives updates.
11. networking components according to claim 5, it is characterised in that described fault detection conversation based on the first and second datum planes are that unidirectional fault detects UFD session.
12. networking components according to claim 11, it is characterised in that described trigger event is the UFD message of the predetermined quantity not received on a described LSP after the fault detection conversation of described first datum plane creates.
13. networking components according to claim 12, it is characterised in that described predetermined quantity is from 1 to 10.
14. in the network system including multiple label switched path with at least one leaf node, a kind of method of transmission switching LSP to the 2nd LSP after breaking down in the first label-switched path LSP, it is characterised in that including:
Create a described LSP;
Creating fault detection conversation based on the first datum plane between the first head node and at least one leaf node described on a described LSP, wherein said fault detection conversation message based on the first datum plane indicates a described LSP to be to activate;
Create described 2nd LSP;
Receive and indicate the instruction broken down of a described LSP, wherein said instruction include not receiving from described first head node at least one described in fault detection conversation message based on the first datum plane;And
By the notification message the second head node from described 2nd LSP that at least one leaf node described is sent to indicate a described LSP to break down.
15. methods according to claim 14, it is characterized in that, further including at and create fault detection conversation based on the second datum plane between the second head node on described 2nd LSP and at least one leaf node described, wherein said fault detection conversation message based on the second datum plane indicates described 2nd LSP to be nonactivated.
16. method according to claim 15, it is characterised in that farther include:
Along described 2nd LSP converting flow with in response to the notification message from described leaf node;And
Sending and indicating described 2nd LSP is that described based on the second datum plane the fault detection conversation message activated is with in response to the notification message from described leaf node.
17. methods according to claim 16, it is characterised in that farther include that handoff notification message is sent to described instruction switching and the first head node after described fault detect based on the first datum plane occurs.
18. methods according to claim 17, it is characterised in that farther including to send the described LSP of instruction is nonactivated described fault detection conversation message based on the first datum plane.
19. methods according to claim 16, it is characterised in that sending a notification message and sending described 2nd LSP of instruction is that the time between described based on the second datum plane the fault detection conversation message activated is less than 200 milliseconds.
20. methods according to claim 14, it is characterised in that farther include that notification message is sent to described instruction and have occurred and that the first head node of switching.
CN201280049788.4A 2011-10-11 2012-10-11 Fault detect in the end-to-end protection solution of multiprotocol label switching multicast label switched path Active CN103891215B (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US201161545897P 2011-10-11 2011-10-11
US61/545,897 2011-10-11
PCT/US2012/059800 WO2013055950A1 (en) 2011-10-11 2012-10-11 Failure detection in the multiprotocol label switching multicast label switched path's end-to-end protection solution

Publications (2)

Publication Number Publication Date
CN103891215A CN103891215A (en) 2014-06-25
CN103891215B true CN103891215B (en) 2016-11-30

Family

ID=

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011006541A1 (en) * 2009-07-16 2011-01-20 Telefonaktiebolaget Lm Ericsson (Publ) Recovery mechanism for point-to-multipoint traffic
WO2011031812A1 (en) * 2009-09-14 2011-03-17 Alcatel-Lucent Usa Inc. Fast upstream source failure detection

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011006541A1 (en) * 2009-07-16 2011-01-20 Telefonaktiebolaget Lm Ericsson (Publ) Recovery mechanism for point-to-multipoint traffic
WO2011031812A1 (en) * 2009-09-14 2011-03-17 Alcatel-Lucent Usa Inc. Fast upstream source failure detection

Similar Documents

Publication Publication Date Title
US7508755B2 (en) Methods and devices for creating an alternate path for a bi-directional LSP
JP5805318B2 (en) Bootstrapping fault detection sessions across P2MP tunnels
CN104067573B (en) A kind of network node of the multicast tag distribution protocol based on point-to-point
EP2087712B1 (en) Method and apparatus for computing alternate multicast/broadcast paths in a routed network
CN101523354B (en) Protecting multi-segment pseudowires
CN102396193B (en) By the method for client data transmission of packet switching provider network
CN104272677B (en) Point to multi-point based multicast label distribution protocol local protection solution
CN106487675A (en) For the outlet protection in EVPN with BUM flow during link failure
JP5691703B2 (en) Multicast network system
US20090225650A1 (en) Path reroute in a computer network
CN101702658B (en) Implement method of ring network protection and system thereof
CN103460647A (en) Technique for operating network node
CN101136788A (en) Fault location method and system for MPLS multicast
CN101815006B (en) aggregation control method of links passing through provider network and system
CN107070788A (en) Pass through the long-range loop free alternate backup path MLDP only quick heavy-routes of multicast
WO2012037820A1 (en) Multi-protocol label switch system, node device and method for establishing bidirectional tunnel
CN101562536A (en) Rapid rerouting method for protecting local path
CN108768796A (en) Link failure detection method and device
CN101155076A (en) Method for checking pseudo wire trouble of two-layer special network
CN104852809B (en) The processing method and system of Signal Degrade failure
CN101529869A (en) Method and apparatus for computing alternate multicast/broadcast paths in a routed network
CN101483491B (en) Shared guard ring, multicast source route protection method and node thereof
EP2767052B1 (en) Failure detection in the multiprotocol label switching multicast label switched path's end-to-end protection solution
CN101599873B (en) Detection method of connectivity and device therefor
CN103891215B (en) Fault detect in the end-to-end protection solution of multiprotocol label switching multicast label switched path

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
GR01 Patent grant