CN101237409A - Method and device for realizing fast rerouting in MPLS VPN - Google Patents

Abstract

The invention discloses a method for realizing fast rerouting in a multi-protocol label switching virtual private network (MPLS VPN), comprising: a border router generates two routes obtained together with labels to be a primary label switching path (Label Switch Path, LSP ) and standby LSP, and send the active LSP and standby LSP to the forwarding table at the same time; check whether the status of the active LSP is valid, if so, use the active LSP to forward traffic, otherwise, use the standby LSP to forward traffic . Correspondingly, the present invention also discloses a routing device. According to a method and device for realizing fast rerouting in MPLS VPN according to the present invention, fast rerouting can be realized, so that services can be quickly converged when a device fails, and real-time Service QoS.

Description

Method and device for implementing fast rerouting in MPLS VPN

technical field

The invention relates to the field of electronic communication, in particular to a method for realizing fast rerouting by a border router in a multi-protocol label switching virtual private network and a routing device.

Background technique

With the development of network technology, multi-protocol label switching (Multi-Protocol Label Switching, MPLS) virtual private network (Virtual Private Network, VPN) is widely used in metropolitan area networks and backbone networks, except for carrying VPN users and large customers It is also used to bear the key services of telecommunications itself such as the third generation mobile communication (3G) and softswitch. Due to the development of more and more services, operators pay more and more attention to VPN cross-domain requirements.

In the MPLS VPN cross-domain application scenario, when problems such as inter-domain link failures and autonomous system boundary router (Autonomous System Boundary Router, ASBR) equipment failures occur, the existing technology for routing convergence includes local ASBR routing convergence and remote ASBR routing convergence. Route convergence.

1. Local ASBR route convergence:

As shown in Figure 1, it illustrates a schematic diagram of ASBR inter-domain routing when VPN services are deployed across domains; in Figure 1, Autonomous Systems (Autonomous System, AS) 1 and AS2 are two MPLS domains. Protocol External Border Gateway Protocol (Multi-Protocol External Border Gateway Protocol, MP-EBGP) (Option B scheme) to achieve cross-domain exchange of VPN routing information. Among them, client edge router (Client Edge, CE) 1 and CE3 both belong to VPN1, and CE3 is dual-homed to two network provider edge routers (Provider Edge, PE) devices, that is, PE3 and PE4 in Figure 1. Both CE2 and CE4 belong to VPN2, and CE4 is dual-homed to two PEs, that is, PE3 and PE4 in Figure 1.

In AS1, establish Interior Border Gateway Protocol (IBGP) neighbor relationships between PE1 and PE2, PE1 and ASBR1, PE2 and ASBR2, and ASBR1 and ASBR2, assuming that ASBR1 and ASBR2 are local ASBR devices; in AS2 , establish IBGP neighbor relationships between PE3 and PE4, PE3 and ASBR3, PE4 and ASBR4, and ASBR3 and ASBR4, assuming that ASBR3 and ASBR4 are remote ASBR devices. Between AS1 and AS2, ASBR1 and ASBR3, and between ASBR2 and ASBR4 establish External Border Gateway Protocol (EGBP) neighbor relationships, and ASBR1 can receive the VPN route advertised by ASBR3, that is, VPN route 1 in Figure 1 At the same time, ASBR1 can also receive the VPN route advertised by ASBR4 through ASBR2, that is, VPN route 2 in FIG. Run the BGP bidirectional forwarding detection mechanism (BFD for BGP) between ASBR1 and ASBR3 to realize real-time detection of equipment and link failures.

Taking VPN1 route as an example, the main process of local ASBR route convergence is as follows: Assume that after traffic arrives from CE1 to ASBR1, ASBR1 forwards traffic across domains using VPN route 1. Figure 2 is a schematic diagram of an existing local ASBR route convergence when VPN services are deployed across domains. As shown in Figure 2, when the ASBR3 device fails or the link between ASBR1 and ASBR3 fails during traffic forwarding, then, ASBR1 can quickly sense this kind of failure through BFD for BGP, and then directly trigger the local convergence of VPN routes, including:

ASBR1 re-optimizes the route, and takes the VPN route 2 published by ASBR4 through ASBR2 as the preferred route; ASBR1 delivers the new preferred route to the forwarding plane, and removes the original route published by ASBR3; after the forwarding table entry of the forwarding plane is updated, the traffic arrives After ASBR1, VPN route 2 will be selected to forward traffic to CE3 across domains, so as to achieve re-convergence of VPN cross-domain services.

However, during the existing local ASBR route convergence process:

After ASBR1 detects a device or link failure within 30ms through BFD, it reports to the interface board, which usually takes about 100ms;

The interface board notifies the main control board to notify the convergence of the BGP protocol, which usually takes several hundred milliseconds to 1 second;

The re-convergence of routes also depends on the busyness of the control plane of ASBR1 and the number of VPN routes. That is, the busy control plane and the large number of VPN routes will reduce the convergence speed of VPN routes to a certain extent.

It can be seen that the existing local ASBR routing convergence speed is very slow, which is unacceptable to users for services with high real-time requirements such as voice and video.

2. Remote ASBR route convergence:

As shown in FIG. 3 , it illustrates a schematic diagram of intra-domain routing of remote ASBRs when the current VPN service is deployed across domains. In Figure 3, autonomous system (Autonomous System, AS) 1 and AS2 are two MPLS domains, during which a single-hop MP-EBGP (Option B scheme) is established to realize cross-domain exchange of VPN routing information. Among them, client edge router (Client Edge, CE) 1 and CE3 belong to VPN1, and CE3 is dual-homed to two network provider edge routers (Provider Edge, PE) devices, namely PE3 and PE4 in the figure. Both CE2 and CE4 belong to VPN2, and CE4 is dual-homed to two PEs, that is, PE3 and PE4 in the figure.

In AS1, establish IBGP neighbor relationships between PE1 and PE2, PE1 and ASBR1, PE2 and ASBR2, and ASBR1 and ASBR2. Assume that ASBR1 and ASBR2 are local ASBRs; Establish an IBGP neighbor relationship among ASBR4, ASBR3, and ASBR4. Assume that ASBR3 and ASBR4 are remote ASBRs. ASBR3 can receive the VPN route advertised by PE3, which is VPN route 3 in the figure. At the same time, ASBR3 can also receive the VPN route advertised by PE4 via ASBR4, which is VPN route 4 in the figure. For the sake of brevity, it is not marked in the figure Any P router. Run BFD for BGP between ASBR3 and PE3 to realize real-time detection of equipment and link faults.

Taking VPN1 route as an example, the main process of remote ASBR route convergence is as follows: Assume that after the traffic reaches ASBR3 across domains from CE1, ASBR3 prefers VPN route 3 to continue forwarding the traffic. As shown in Figure 4, when PE3 fails or the link between ASBR3 and PE3 fails during traffic forwarding, ASBR3 can quickly detect the failure through BFD for BGP, and then directly trigger local convergence of VPN routes. include:

ASBR3 re-optimizes routes, using VPN route 4 published by PE4 through ASBR4 as the preferred route; ASBR3 sends the new preferred route to the forwarding plane, and removes the original VPN route 3 published by PE3; after the forwarding table entry of the forwarding plane is updated, After the traffic reaches ASBR3 across domains, VPN route 4 is selected to forward the traffic to CE3, so as to achieve re-convergence of VPN services.

However, during the existing remote ASBR route convergence process:

After ASBR3 detects a device or link failure within 30ms through BFD, it reports to the interface board, which usually takes about 100ms;

The interface board notifies the main control board to notify the convergence of the BGP protocol, which usually takes several hundred milliseconds to 1 second;

The re-convergence of routes also depends on the busyness of the ASBR3 control plane and the number of VPN routes. That is, if the control plane is busy and the number of VPN routes is large, the convergence speed of VPN routes will be reduced to a certain extent.

It can be seen that the routing convergence speed of the existing remote ASBR is very slow, which is unacceptable to users for services with high real-time requirements such as voice and video.

To sum up, in the process of realizing the present invention, the inventor found that in the MPLS VPN cross-domain application scenario, when problems such as inter-domain link failures and ASBR equipment failures occur, the ASBR route convergence speed is relatively slow, and for real-time services , seriously affecting its quality of service (Quality of Service, QoS), unable to meet user needs.

Contents of the invention

The embodiment of the present invention is to provide a method and routing equipment for implementing fast rerouting in a multi-protocol label switching virtual private network (MPLS VPN), which can realize fast rerouting, achieve fast convergence of services when equipment fails, and improve real-time service reliability. QoS.

In order to achieve the above-mentioned technical effects, the embodiment of the present invention provides a method for realizing fast rerouting in MPLS VPN, which includes:

Border routers receive traffic;

Query the forwarding table to obtain the state of the active LSP corresponding to the traffic, and store the information of the active Label Switch Path (Label Switch Path, LSP) and standby LSP and the state of the active LSP in the forwarding table;

If the state of the active LSP is valid, use the active LSP to forward the traffic; or if the state of the active LSP is invalid, use the standby LSP corresponding to the traffic to forward the traffic.

Correspondingly, the embodiment of the present invention also provides a routing device, which includes:

The forwarding table storage module is used for storing and storing the forwarding table, wherein the forwarding table stores the information of the active LSP, the standby LSP, and the state of the active LSP;

A query module, configured to query whether the outgoing interface of the active LSP is valid;

The LSP selection module is used to select the active LSP to forward traffic when the query module finds that the state of the active LSP is valid, and select the standby LSP to forward traffic when the query module finds that the outgoing interface of the active LSP is invalid.

According to the method and routing device for implementing fast rerouting in an MPLS VPN proposed by the embodiment of the present invention, two routes are generated as active and standby LSPs, and are installed in the forwarding table at the same time, and the forwarding is selected by querying the state of the active LSP link, when the active LSP fails, immediately switch the traffic to the standby LSP, effectively realizing fast rerouting, improving convergence speed, ensuring QoS, and better meeting user needs.

Description of drawings

Figure 1 is a schematic diagram of ASBR inter-domain routing when the current VPN service is deployed across domains;

Figure 2 is a schematic diagram of an existing local ASBR route convergence when VPN services are deployed across domains;

Figure 3 is a schematic diagram of remote ASBR intra-domain routing when the current VPN service is deployed across domains;

Figure 4 is a schematic diagram of an existing remote ASBR route convergence when VPN services are deployed across domains;

Fig. 5 is the flow chart of route configuration in the method for realizing fast rerouting in a kind of MPLS VPN that the embodiment of the present invention proposes;

Fig. 6 is a flow chart of route selection in a method for realizing fast rerouting in an MPLS VPN proposed by an embodiment of the present invention;

Fig. 7 is a flow chart of routing convergence in a method for realizing fast rerouting in an MPLS VPN proposed by an embodiment of the present invention;

Fig. 8 is an application scenario diagram of a method for implementing fast rerouting in an MPLS VPN in an ASBR inter-domain according to an embodiment of the present invention;

Fig. 9 is an application scenario diagram in an ASBR domain of a method for realizing fast rerouting in an MPLS VPN according to an embodiment of the present invention;

FIG. 10 is a schematic structural diagram of a routing device proposed by an embodiment of the present invention;

FIG. 11 is a schematic structural diagram of a forwarding table of an ASBR in an embodiment of the present invention;

Fig. 12 is a schematic structural diagram of the forwarding table of PE1 in the embodiment of the present invention.

Detailed ways

A method for realizing fast rerouting in an MPLS VPN proposed by an embodiment of the present invention generates active and standby Label Switch Paths (Label Switch Path, LSP) by generating two routes, and sends them to forwarding table entries at the same time, Select the forwarding link by querying the state of the active LSP. When the active LSP fails, immediately switch the traffic to the standby LSP to achieve fast rerouting. The specific implementation includes at least: routing configuration, routing selection when forwarding traffic, Route convergence in the event of a failure.

Referring to Fig. 5, it illustrates a flow chart of routing configuration in a method for implementing fast rerouting in an MPLS VPN according to an embodiment of the present invention, specifically including:

Step S11, the border routing device receives at least two VPN routes, one VPN route is used as the main route, and one VPN route is used as the backup route;

Step S12, the active route and the standby route together with labels (including inter-domain labels or intra-domain labels, wherein the inter-domain labels or intra-domain labels include outgoing labels and incoming labels) generate active LSPs and standby LSPs respectively, wherein , one incoming label corresponds to two primary and secondary LSPs;

Step S13, install the active LSP and the standby LSP into the MPLS forwarding table of the border routing device at the same time, wherein, the forwarding table is shown in Figure 11, and its entries include at least: incoming label, outgoing label, next Hop, outgoing interface, and active LSP status.

Step S14, running BFD on the active LSP to perform link detection.

When the border router receives the MPLS message and needs to forward the traffic (or message), it selects the route, as shown in Figure 6, which illustrates the realization of fast rerouting in a kind of MPLS VPN proposed by the embodiment of the present invention The flow chart of route selection in the method, including:

Step S21, query the status of the primary LSP in the MPLS forwarding table according to the label carried by the traffic, specifically, in conjunction with Figure 11, look up the incoming label corresponding to the label carried by the traffic in the forwarding table according to the label carried by the traffic, so as to obtain The state of the active LSP corresponding to the incoming label;

Step S22, judging whether the state of the active LSP is valid, if the judging result is yes, go to step S23, otherwise go to step S24;

Step S23, selecting the primary LSP to forward traffic;

Step S24, selecting an alternate LSP to forward traffic.

In this embodiment, the traffic forwarding path is selected by judging the status of the active LSP, and when the active LSP is unavailable, the traffic can be switched to the standby LSP in time to achieve fast convergence.

When in the process of traffic forwarding, in step S14, a link failure or device failure is detected through BFD, and route convergence needs to be performed, as shown in Figure 7, including:

Step S31, the BFD detects a failure of the active LSP, including: link failure or equipment failure;

Step S32, modifying the state of the active LSP in the MPLS forwarding table to be invalid, that is: setting the state of the active LSP to down, and reporting to the control layer;

Step S34, in the traffic forwarding process, as shown in Figure 6, by querying the MPLS forwarding table, it is found that the state of the active LSP is down, that is: invalid;

Step S35, selecting a backup LSP for traffic forwarding.

By running BFD, ensure that fault detection is completed within 30 ms. Once a fault is detected on the active LSP, the state of the active LSP in the forwarding table will be updated in time to implement fast rerouting and ensure fast convergence within 50 ms. This improves the convergence speed and reduces the time delay, improve the user experience of real-time services, and ensure the quality of service.

In order to further elaborate a kind of method of realizing fast rerouting in MPLS VPN of the present invention, below in conjunction with accompanying drawing, illustrate from two aspects of fast rerouting (Fast ReRoute, FRR) between ASBR domain and FRR in ASBR domain respectively below.

Referring to FIG. 8 , it illustrates an application scene diagram of a method for implementing fast rerouting in an MPLS VPN in an ASBR inter-domain according to an embodiment of the present invention.

In the figure, AS1 and AS2 are two MPLS domains, and a single-hop MP-EBGP (Option B scheme) is established between them to realize cross-domain exchange of VPN routing information. Among them, CE1 and CE3 belong to VPN1, and CE3 is dual-homed to two PE devices, that is, PE3 and PE4 in the figure. Both CE2 and CE4 belong to VPN2, and CE4 is dual-homed to two PEs, that is, PE3 and PE4 in the figure. In AS1, establish IBGP neighbor relationships between PE1 and PE2, PE1 and ASBR1, PE2 and ASBR2, and ASBR1 and ASBR2. Assume that ASBR1 and ASBR2 are local ASBRs; Establish an IBGP neighbor relationship among ASBR4, ASBR3, and ASBR4. Assume that ASBR3 and ASBR4 are remote ASBRs. Between AS1 and AS2, ASBR1 and ASBR3, and between ASBR2 and ASBR4 respectively establish EGBP neighbor relationships. ASBR1 can receive the VPN route advertised by ASBR3, that is, VPN route 1 in the figure. At the same time, ASBR1 can also receive ASBR4 passing through ASBR2. The advertised VPN route is VPN route 2 in the figure. To simplify the description, no P router is marked in the figure. Run BFD for BGP between ASBR1 and ASBR3 to realize real-time detection of equipment and link failures.

In the embodiment of the present invention, ASBR1 usually selects the VPN route 1 published by ASBR3 as the primary route, and selects the VPN route 2 published by ASBR4 as the backup route, and combines it with the label in the AS1 domain to generate the primary LSP1 and backup LSP2 respectively. ASBR1 delivers the active LSP1 and standby LSP2 to the MPLS forwarding table at the same time.

Under normal circumstances, after traffic arrives at ASBR1 from CE1, it forwards cross-domain traffic to ASBR3 through LSP1. When ASBR3 fails or the link between ASBR1 and ASBR3 fails, and ASBR1 quickly detects it through BFD, it immediately triggers inter-domain FRR. , switch the traffic to the standby LSP2, that is, after the traffic arrives at ASBR1 from CE1, it forwards the cross-domain traffic to CE3 through LSP2, thereby ensuring the 50ms fast switching of the traffic and greatly improving the convergence speed of the cross-domain VPN service.

Wherein, triggering inter-domain FRR, and switching the traffic to the standby LSP2 is specifically: setting the status of the active LSP1 in the forwarding table to down, and when the traffic arrives, it is found that the status of the active LSP1 is down by querying the forwarding table, then Immediately enable the standby LSP2 in the MPLS forwarding table for traffic forwarding.

Referring to FIG. 9, it illustrates an application scene diagram of a method for implementing fast rerouting in an MPLS VPN in an ASBR domain according to an embodiment of the present invention.

In the figure, AS1 and AS2 are two MPLS domains, and a single-hop MP-EBGP (Option B scheme) is established between them to realize cross-domain exchange of VPN routing information. Among them, CE1 and CE3 belong to VPN1, and CE3 is dual-homed to two PE devices, that is, PE3 and PE4 in the figure. Both CE2 and CE4 belong to VPN2, and CE4 is dual-homed to two PEs, PE3 and PE4 in the figure. In AS1, establish IBGP neighbor relationships between PE1 and PE2, PE1 and ASBR1, PE2 and ASBR2, and ASBR1 and ASBR2. Assume that ASBR1 and ASBR2 are local ASBRs; ASBR4, ASBR3, and ASBR4 respectively establish IBGP neighbor relationships. Assuming that ASBR3 and ASBR4 are remote ASBR devices, ASBR3 can receive the VPN route advertised by PE3, that is, VPN route 3 in the figure. At the same time, ASBR3 can also receive the The VPN route advertised by ASBR4 is VPN route 4 in the figure. For the sake of brevity, no P router is marked in the figure. Run BFD for BGP between ASBR3 and PE3 to realize real-time detection of equipment and link faults.

In the embodiment of the present invention, ASBR3 usually selects the VPN route 3 published by PE3 as the active route, and selects the VPN route 4 published by PE4 as the backup route, and combines them with the inter-domain label to generate the active LSP3 and the backup LSP4 respectively. ASBR3 delivers the active LSP3 and standby LSP4 to the MPLS forwarding table at the same time.

Under normal circumstances, after the traffic reaches ASBR3 across domains from CE1, it forwards the traffic to CE3 through LSP3. When PE3 fails or the link between ASBR3 and PE3 fails, and ASBR3 quickly detects it through BFD, it immediately triggers the intra-domain FRR, switch the traffic to the standby LSP4, that is, after the traffic reaches ASBR3 across the domain from CE1, forward the cross-domain traffic to CE3 through LSP4, thus ensuring the 50ms fast switching of the traffic and greatly improving the convergence speed of the cross-domain VPN service .

Wherein, triggering the FRR in the domain, and switching the traffic to the standby LSP4 is specifically: setting the state of the active LSP3 in the forwarding table to down, and when the traffic arrives, it is found that the state of the active LSP3 is down by querying the forwarding table, then Immediately enable the backup LSP4 in the MPLS forwarding table for traffic forwarding.

The above-mentioned embodiments of the present invention show a scheme of configuring FRR between ASBRs or within an ASBR. Of course, a scheme of configuring FRR between ASBRs and configuring FRR within an ASBR may also be used.

In one embodiment of the present invention, on the basis of configuring FRR between ASBR domains and/or within domains, VPN FRR can be further configured on PE1, wherein the forwarding table of PE1 is shown in Figure 12, and its entries include: VPN route prefix, VPN route, private network label, public network label, next hop, outbound interface, and LSP status corresponding to the active VPN route;

Referring to Figure 8 or Figure 9, under normal circumstances, when receiving traffic sent by CE1, PE1 selects the active LSP corresponding to PE1-ASBR1 to forward the traffic,

When a PE1-ASBR1 link failure or ASBR1 equipment failure occurs, to ensure fast convergence of VPN services within 50 ms, PE1 selects the backup LSP corresponding to PE1-PE2-ASBR2 for traffic forwarding, which implements fast rerouting on PE1 and ensures Reliability within AS1, combined with ASBR inter-domain and intra-domain FRR in the entire MPLS VPN, can achieve end-to-end protection and improve the reliability and disaster recovery of the entire system.

Based on the above-mentioned method of implementing fast rerouting in MPLS VPN, to realize fast convergence when ASBR and its links fail, it is necessary to expand the functions of the corresponding equipment.

As shown in Figure 10, an embodiment of the present invention proposes a routing device, which includes:

The forwarding table storage module 1020 is used to store the forwarding table, and the forwarding table stores the information of the active LSP, the standby LSP, and the state of the active LSP; wherein, when the border router is an ASBR, for example, the structure of the forwarding table As shown in FIG. 11 , its entries at least include: incoming label, outgoing label, next hop, outgoing interface, and state of the active LSP.

The query module 1030 is used to query whether the status of the active LSP is valid according to the label carried by the traffic when forwarding traffic. Specifically, referring to FIG. Ingress label, so as to obtain the status of the active LSP corresponding to the ingress label;

The LSP selection module 1040 is configured to select the active LSP to forward traffic when the query module 1030 finds that the active LSP is valid, and select the standby LSP to forward traffic when the active LSP is invalid.

In another embodiment of the present invention, the routing device further includes:

The LSP generating module 1010 is configured to respectively generate the active LSP and the standby LSP together with the obtained two routes together with the labels.

In another embodiment of the present invention, the routing device further includes:

Fault detection module 1050, configured to detect whether the active LSP fails;

The fault processing module 1060 is configured to set the state corresponding to the active LSP in the forwarding table storage module 1020 to an invalid state when the fault detection module 1050 detects that the active LSP fails, and report to the control layer.

It should be noted that, the routing device described in the embodiment of the present invention may specifically include an autonomous system border router (ASBR) or a network provider (PE) border router.

In the above-mentioned embodiment, the method and equipment for realizing fast rerouting of the present invention are illustrated by using BFD to realize fault detection and modifying the state of the active LSP as an example. Of course, the present invention is not limited to this, and Ethernet operation and management and maintenance (Operations, Administration and Maintenance, OAM); if the LSP uses the packet (Packet OverSONET/SDH) on the Synchronous Optical Network (Synchronous Optical Network, SONET)/Synchronous Digital Hierarchy (SDH), POS) link, usually through the POS self-alarm mechanism, realizes fault detection, and jointly modifies the state of the active LSP.

In summary, according to the embodiment of the present invention, two routes are generated as active and standby LSPs and installed in the forwarding entry at the same time, and the forwarding link (main LSP or standby LSP) is selected by querying the state of the active LSP. ), and run BFD between inter-domain ASBRs and between intra-domain ASBRs and PEs to implement bidirectional forwarding detection. Update the forwarding table, or use the self-tightening mechanism of the POS link or the OAM mechanism to realize fast fault detection, and update the forwarding table in time to speed up the triggering of FRR and switch the traffic forwarding to the standby LSP to ensure fast convergence within 50 ms. It greatly improves the reliability of VPN across domains, ensures QoS, and can better meet user needs.

The above description is a preferred embodiment of the present invention, and it should be pointed out that for those skilled in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also considered Be the protection scope of the present invention.

Claims (11)

1, a kind of method that realizes quick heavy-route in MPLS VPN is characterized in that, comprising:

Border router receives flow;

Inquiry is transmitted, and obtains the main state with label switching path LSP of described flow correspondence, has stored the first main information and the first main state with LSP with the LSP and the first standby LSP in described the transmitting;

If the described first main state with LSP is effectively, then use described first master to transmit described flow with LSP; Or

If the described first main state with LSP is invalid, then use the first standby LSP converting flow of described flow correspondence.

2, the method for claim 1 is characterized in that, also is included in the step that configuration is transmitted on the described autonomous border router, is specially:

Two routes that border router will obtain generate the first main LSP of using and the first standby LSP respectively together with label;

Described first main being issued to the LSP and the first standby LSP transmitted.

3, method as claimed in claim 2 is characterized in that, the described list item of transmitting comprises at least: go into label, outgoing label, next jumping, outgoing interface, main state with LSP.

4, the method for claim 1 is characterized in that, also comprises:

Main move BFD on LSP described first;

Described first main when use lsp failure when detecting by BFD, with described first lead state with LSP be changed to invalid.

5, the method for claim 1 is characterized in that, when the grouping POS link on described LSP employing Synchronous Optical Network/synchronous digital transmission system, described method also comprises:

Whether detect described first master by POS self alarming mechanism breaks down with LSP;

Described first main when use lsp failure when detecting, with described first lead state with LSP be changed to invalid.

6, as any described method of claim 1 to 4, it is characterized in that described inquiry is transmitted, obtain the first main state of described flow correspondence, comprising with LSP:

In described transmitting, search the main state with LSP corresponding according to the label that flow carries with this label.

7, method as claimed in claim 6 is characterized in that, described method further comprises:

Two routes that the network provider border router will obtain generate the second main LSP of using and the second standby LSP respectively together with label;

Described second master is issued in the transmitting of network provider border router with the LSP and the second standby LSP;

When flow was transmitted, whether the inquiry second main state with LSP was effective, if, then use the second main LSP converting flow of using, otherwise, use the second standby LSP to carry out converting flow.

8, a kind of routing device is characterized in that, comprising:

Transmit memory module, be used for forwarding table memory, preserved main in described the transmitting with LSP, standby LSP information and main state with LSP;

Enquiry module, whether the state that is used for the described main LSP of using of inquiry when flow is transmitted is effective;

LSP chooses module, is used for inquiring main state with LSP when effective when enquiry module, chooses the main LSP converting flow of use, when enquiry module inquires the state of the main LSP of using when invalid, chooses standby LSP converting flow.

9, routing device as claimed in claim 8 is characterized in that, also comprises:

The LSP generation module is used for generating described two routes described main with LSP and described standby LSP respectively together with described label.

10, routing device as claimed in claim 8 or 9 is characterized in that, also comprises:

Whether fault detection module is used to detect described master and breaks down with LSP;

Fault processing module is used for when described fault detection module detects described master and breaks down with LSP, transmits with described that main state with the LSP correspondence is changed to disarmed state described in the memory module, and reports key-course.

11, routing device as claimed in claim 10 is characterized in that, described routing device comprises Autonomous System Boundary Router, AS Boundary Router.

Images