WO2011144088A2

WO2011144088A2 - Method for service protection and access device

Info

Publication number: WO2011144088A2
Application number: PCT/CN2011/074683
Authority: WO
Inventors: 江元龙; 罗勇
Original assignee: 华为技术有限公司
Priority date: 2011-05-26
Filing date: 2011-05-26
Publication date: 2011-11-24
Also published as: CN102282805A; CN102282805B; WO2011144088A3

Abstract

A method for service protection and an access device are provided. The method includes: the continuity fault detection of service level is performed at a service path between an access node and aggregation node, and a maintenance end point for performing the continuity fault detection of service level is configured at the access node and aggregation node; when the access node detects the continuity fault detection of service level, a standby physical link of a main physical link is activated, so that the standby Label Switching Path (LSP) and standby pseudo wire of the standby physical link are available; the service with fault is switched from the main pseudo wire carrying the service to the standby pseudo wire, wherein the main pseudo wire is the pseudo wire carried by the main LSP at the main physical link, and the standby pseudo wire is the pseudo wire carried by the standby LSP at the standby physical link. By using the method, the service protection process is simplified, and the protection methods of various services such as unicast, multicast and so on are integrated.

Description

Service protection method and access device

The present invention relates to the field of communications technologies, and in particular, to a service protection method and an access device.

Background technique

MPLS (Multi-Protocol Label Switching) technology is a packet-switching-based packet transmission technology that supports multiple services, provides good traffic engineering capabilities, and flexibility for network deployment. MPLS mainly has two modes of carrying Ethernet services: one is a point-to-point Ethernet private leased line service, and the other is a multi-point to multi-point Ethernet virtual LAN service, which respectively corresponds to VP WS (Virtual Private Wire Service, Virtual private leased line service) and VPLS (Virtual Private LAN Service) technology.

In VPWS, the service packet is added to the MPLS network in the label switching mode after the PW (Pseudowire) label and the LSP (Label Switching Path) tunnel label are added to the ingress node of the MPLS network. Export node. Each of the pseudowires corresponds to a unique PW tag value, and multiple pseudowires can be multiplexed into one LSP tunnel for transmission to improve network scalability. During the transmission of the service through the pseudowire, only the ingress PE (the Provider Edge) and the egress PE node need to process the PW label and perform the service processing. The intermediate node only performs the LSP label exchange, and does not need to process the PW label. . Between each pair of neighbors of the path, the LSP is carried over a physical link layer (e.g., an Ethernet link layer). VPWS is commonly used for the delivery of unicast traffic.

Under VPLS, the MPLS network connects multiple geographically dispersed customer sites by emulating an Ethernet broadcast domain between PE nodes. A VPLS instance consists of multiple PEs. The PEs are connected to each other through pseudowires (pseudowires are also carried on the LSP), and the services are transmitted through the pseudowires. The PE initiates and terminates the pseudowires. The forwarder forwards according to the MAC address of the Ethernet service frame. Since there is no need to replicate each multicast packet at the source node, VPLS is more efficient for multicast traffic transmission than VPWS bandwidth, and is therefore more used for multicast services.

For a packet network that combines the above two technologies for single/multicast multi-service convergence, different protection mechanisms need to be adopted for unicast and multicast services, and multicast services are also segmented. Protection, the protection methods involved are cumbersome and have limited limitations, for unicast The protection mechanism of the multicast service is also not uniform.

Summary of the invention

The embodiments of the present invention provide a service protection method and an access device, which can simplify the service protection process and unify the protection methods under various services such as unicast and multicast.

In order to solve the above technical problem, the technical solution of the embodiment of the present invention is as follows:

A method of business protection, including:

Performing service level continuity fault detection on the service path between the access node and the sink node, where the access node and the sink node are configured with a maintenance endpoint that can perform the service level continuity fault detection;

When the access node detects a continuity fault of the service level, activating the standby physical link of the active physical link, so that the alternate label switched path LSP path and the backup pseudowire on the standby physical link are available;

The faulty service is switched to the standby pseudowire by the primary pseudowire carrying the service, where the primary pseudowire is a pseudowire carried by the primary LSP path on the active physical link. The standby pseudowire is a pseudowire carried by a backup LSP path on the standby physical link.

An access device includes:

a detecting unit, configured to run a service level continuity fault detection on a service path between the access node and the aggregation node, where the maintenance node configured to perform the service level continuity fault detection is configured;

An activation unit, configured to activate a standby physical link of the active physical link when a continuity fault of the service level is detected, so that the backup LSP path and the backup pseudowire on the standby physical link are available; And the faulty service is switched by the primary pseudowire carrying the service to the standby pseudowire, where the primary pseudowire is carried by the primary LSP path on the active physical link. The pseudowire is the pseudowire carried by the alternate LSP path on the standby physical link.

In the embodiment of the present invention, the maintenance endpoint is configured in the access node and the aggregation node of the service path in the single/multicast service, and the service level continuity fault detection is performed between the maintenance endpoints, thereby ignoring different layers in the path. Different MPLS OAM mechanisms between different LSP segments, as long as service level continuity faults are detected, both the unicast and multicast services are performed by the access node to the standby physical link. Activating, so that the backup LSP path carried by the standby physical link and the alternate pseudowire carried by the backup LSP path are available, and the service pseudowire carrying the service is switched from the primary pseudowire on the primary LSP path to the backup LSP path. The backup pseudowire is used to implement service protection under the MPLS multi-service access system. This method not only greatly simplifies the service protection process, but also unifies the protection methods of unicast and multicast services. In addition, the method does not need to rely on the intermediate node for service protection, and can protect multiple fault scenarios of multiple services, ensuring the independence of the access device, and reducing the difficulty of the service access configuration, and does not need to separately unicast and group. Broadcasting provides different protection mechanisms, making deployment easier.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present invention, and other drawings can be obtained from those skilled in the art without any inventive labor.

1 is a schematic diagram of an application scenario of a unicast service in the prior art;

2 is a schematic diagram of an application scenario of a multicast service in the prior art;

Figure 3 is a schematic diagram of a failure of a multicast service;

4 is a flowchart of a service protection method according to an embodiment of the present invention;

Figure 5 is a schematic diagram of switching protection in the embodiment shown in Figure 4;

6 is a flowchart of another service protection method according to an embodiment of the present invention;

7 is a schematic diagram of an application scenario in the embodiment shown in FIG. 6;

8 is a schematic diagram of handover protection in the case of homology and sibling in the embodiment shown in FIG. 6;

9 is a flowchart of another service protection method according to an embodiment of the present invention;

FIG. 10 is a schematic diagram of an application scenario in the embodiment shown in FIG. 9;

FIG. 11 is a schematic structural diagram of another access node and a sink node according to an embodiment of the present invention; FIG. 12 is a schematic structural diagram of an access device according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of another access device according to an embodiment of the present invention;

FIG. 14 is a schematic structural diagram of an MPLS multi-service access system according to an embodiment of the present invention.

detailed description

The detailed description of the present invention and the accompanying drawings are to be understood by the accompanying drawings Bright.

In the prior art, based on the VPWS and VPLS technologies, there are two typical architectures for the MPLS access network. As shown in Figure 1, the access node (AN) AN1 provides a unicast service scenario through the VPWS technology, such as the LSP path of the residential broadband Internet service through the access node AN1 to the Aggregation Nodes (AGS) AGS2. A PW on the top is transmitted. Further, in order to improve the reliability of the service, the access node AN1 is respectively connected to the two nodes AGS1 and AGS3 through the primary and backup physical links, so that two can be established between the access node AN1 and the aggregation node AGS2. The physically separated LSP path, such as the primary LSP path through AN1-AGS1-AGS2, and the alternate LSP path through AN1-AGS3-AGS2, because the primary and backup LSP paths are homogenous and identical, so the same pseudowire can be shared. That is, the service pseudowires between the foregoing AN1 and AGS2 may be carried by the primary LSP path or by the backup LSP path. Of course, for the case of the same-same different sinks, the respective LSP paths carry different pseudowires. Here, the homologous and homogenous representations are imported into the same sink node by the same access node, and the homologous different sinks indicate that the same access node joins different sink nodes. The primary LSP path is carried by the primary physical link of AN1-AGS1 and another one or more physical links between AGS1 and AGS2. The alternate LSP path is the alternate physical link of AN1-AGS3 and another segment between AGS3 and AGS2. Or multiple segments of physical link bearers. When the primary physical link is faulty, it is obvious that the LSP path is faulty. In this case, you need to initiate the switching of the LSP path or switch the LSP path to ensure the normal operation of the service.

As shown in FIG. 2, a scenario in which the access node AN2 provides a multicast service through the VPLS technology. In order to improve the reliability of the service, the access node AN2 is respectively connected to the two nodes AGS4 and AGS6 through two primary and backup physical links, so that there are also two between the access node AN2 and the aggregation node AGS5. The physical separation of the service path, such as the primary path through AN2-AGS4-AGS5, and the alternate path through AN2-AGS6-AGS5, but in the multicast service, the downlink Ethernet service packet provided by BRAS2 is first performed in AGS5. The pseudowire is encapsulated and carried by LSP2, and then the multicast packet is transmitted to the AGS4 through the VPLS network. The AGS4 receives and decapsulates the Ethernet frame and then forwards it according to the MAC address in the Virtual Switch Instance (VSI) repeater module. Finally, the LSP1 arrives at the AN2, so the primary path and the alternate path both include two LSP paths. For example, the primary path includes the LSP1 and LSP2 paths, and the alternate path includes the LSP3 and LSP4 paths. Each segment of the LSP path carries its own corresponding pseudowire, between the access node AN2 and the aggregation node AGS5. The service is carried by the pseudo-line carried by the LSPs in the main path. Each LSP is carried by its corresponding physical layer link. For example, the LSP1 path is carried by the physical layer link of AN1-AGS4. In the multicast service, the AN2-AGS4 physical layer link carrying the LSP1 path and the AN2-AGS6 physical layer link carrying the LSP3 path cannot be activated at the same time. Otherwise, loopback occurs, that is, the downlink multicast packet is After the LSP1 path of the AGS4-AN2 link enters the AN2, the LSP3 path of the other AN2-AGS6 link may be returned to the VPLS network and continue to reach the same AGS device through the VPLS network. Causes a loopback of the path. Therefore, for a multicast service, only the primary physical link is activated. When it fails, it is obvious that the LSP1 it carries will be faulty. In this case, it can be based on the physical layer protocol (such as the Ethernet cross-frame LAG protocol). 802.1AX) activates the standby physical link segment of the active physical link, and initiates the switching of LSP1 to the LSP3 carried by the standby physical link segment and the pseudowire switching of the corresponding LSP segment to ensure normal operation of the service.

When the unicast and multicast services are provided on the same MPLS network, the MPLS OAM needs to be run between the AN and the AGS and the AGS device in the above two scenarios. The Administration and Maintenance mechanism is used to detect whether there is a fault in the corresponding LSP path and the LSP path segment. For example, the LSP path between AN1-AGS2 in Figure 1 and the LSP path between AN2-AGS4 in Figure 2 LSP path between AGS4-AGS5, etc. As the VPLS edge node is used as the VPLS edge node, the LSPs need to be terminated and forwarded through the Ethernet destination address. The MPLS OAM running between different LSP segments is completely independent. If a certain LSP fails, the endpoints of other LSP segments are usually incapable of sensing. For example, Figure 3 In the case that the LSP2 path between the AGS4-AGS5 fails, the AN2 cannot detect the fault if it relies on the LSP path layer OAM. Therefore, when performing service protection, the unicast and multicast services can only be separately protected, and The multicast service also needs to perform segmented LSP protection, and the segmentation protection also involves the synchronization problem between the intermediate dual-homed device nodes, so the protection action is very complicated. Therefore, the segmentation protection method in the prior art is cumbersome and has limited limitations, and the protection mechanism for the unicast and multicast services is not uniform.

Based on this, for a packet network that combines the single/multicast multi-service convergence of the above two technologies, the embodiment of the present invention provides a service protection method and an access device, in a single/multicast service, in a service path. Configure maintenance endpoints at the access nodes and aggregation nodes, and perform service level between maintenance endpoints. Continuity fault detection, which eliminates the need to deploy corresponding MPLS OAM mechanisms and physical layer OAMs between different layers and different LSP segments in the path. As long as service level continuity faults are detected, both unicast and multicast services are available. The activation of the backup physical link is performed by the access node, so that the backup LSP path carried by the standby physical link and the backup pseudowire carried by the backup LSP path are available, and the service pseudowire carrying the service is used on the primary LSP path. The primary pseudowire switches to the alternate pseudowire of the alternate LSP path, thereby implementing service protection under the MPLS multi-service access system. This method not only greatly simplifies the service protection process, but also unifies the protection methods of unicast and multicast services.

The technical solution of the present invention will be described below with reference to the accompanying drawings and embodiments.

FIG. 4 is a flowchart of a service protection method according to an embodiment of the present invention.

In this embodiment, the service transmitted between the access node and the sink node may be a unicast service or a multicast service. The access node may be a digital subscriber line access multiplexer when the copper wire is accessed.

(Digital Subscriber Line Access Multiplexer, DSLAM), which may be an Optical Line Terminal (OLT) when optical access is used, or an MPLS-enabled access device that uses other access technologies or multiple access technologies. The aggregation node can be an MPLS-enabled service router, such as the Quidway NE40 full-service router.

Before the fault occurs, the service is carried by the primary physical link between the access node and the sink node. Specifically, the service is carried by the primary pseudowire carried by the primary LSP path on the primary physical link. In order to facilitate the subsequent service protection switching, the physical link between the access node and the sink node also has a standby physical link of the active physical link, and the backup physical link also carries the corresponding backup LSP path and the backup pseudo. Line (as mentioned above, the alternate pseudowire and the main pseudowire can be the same when homologous and sibling).

In this embodiment, a maintenance end point (MEP) is configured on the access node and the aggregation node, for example, on the access node AN1 and the aggregation node AGS2 shown in FIG. 1, or in FIG. A maintenance endpoint is configured on the access node AN2 and the aggregation node AGS5, and the maintenance endpoint can send, receive, and process OAM messages such as service level continuity fault detection messages and reverse fault notifications, where the service level continuity fault detection message is For example, it is transmitted in a pseudowire between the access node and the sink node, and the service running between the access node and the sink node may be an Ethernet service, and between the two maintenance endpoints, such as Y.1731 or 802.1ag OAM may be enabled. Service level continuity fault detection protocol. The two maintenance endpoints may be functional modules respectively built in the access node and the aggregation node, or may be respectively disposed outside the access node and the aggregation node and connected thereto. The physical mode is no longer - enumerated, as long as the service level continuity fault detection on the primary path can be performed through the two maintenance endpoints. Among them, the service-level continuity fault detection is any protocol specification that adopts the Ethernet header encapsulation for the Ethernet service in accordance with Y.1731, 802. lag or other similar OAM functions, and the prior art is in different segments of the LSP path. The path layer OAM performed internally is MPLS/PW OAM, such as Bidirectional Forwarding Detection (BFD), which uses the MPLS header encapsulation. In addition, for a unicast service, a point-to-point service OAM is usually used. The source address of the downlink OAM message is the unicast MAC address of the sink node, and the destination address is the unicast MAC address of the access node. For a multipoint OAM (the source address in the downlink OAM message is the unicast MAC address of the sink node, and the destination address is the multicast MAC address). After the service endpoint is configured, the service protection method in the MPLS multi-service access system may include the following steps:

Step 401: Run a service level continuity fault detection on a service path between the access node and the sink node.

In this step, a service level continuity fault detection message (CCM) may be periodically sent from the maintenance endpoint MEP of the sink node to the maintenance node MEP of the access node, and then the access node MEP is based on the scheduled time. Whether the CCM message of the other party is received (for example, 3.5 times the above period) to verify whether the operation of a certain service is faulty. If a service failure occurs, it indicates that a certain physical chain carrying the service between the access node and the aggregation node A primary physical link such as an access node that is dual-homed, or another physical link on the path, or a node on the path may fail. The service level continuity fault detection can also be:

The access node sends a service level continuity fault detection message to the sink node;

If the service-level continuity fault detection response message returned by the aggregation node is received within the preset time, the service is running normally. Otherwise, if the service-level continuity fault detection response message sent by the aggregation node is not received within the preset time, The fault indicates that the service is faulty in the downlink direction from the sink node to the access node. If the reverse fault indication message sent by the sink node is received, it indicates that the service is faulty in the uplink direction from the access node to the sink node. .

Step 402: When the access node detects a service level continuity detection fault, activate the standby physical link of the active physical link, so that the standby LSP path and the backup pseudowire on the standby physical link are available.

Service-level continuity fault detection messages running on the two maintenance endpoints contain service information. When the ingress node detects a service running fault, the access node relinquishes the primary physical link carrying the service and activates the standby physical link, so as to switch the service to the backup LSP path and the alternate pseudo of the standby physical link. In the embodiment of the present invention, the unicast or multicast service, before the service level continuity fault detection is detected, the standby physical link is inactive, and the standby physical link is activated when the fault is detected. The activation process may be the activation of a physical port, or only logically allow the transmission and processing of service data frames, and will not be described again in the future.

Step 403: Switch the faulty service from the primary pseudowire carrying the service to the standby pseudowire. After the standby physical link is activated, the standby LSP path and the backup pseudowire on the normal physical link can transmit the service data frame. The access node can switch the faulty service from the primary pseudowire carrying the service to the corresponding standby pseudo. After the line, the service is transmitted through the alternate pseudowire carried by the standby LSP on the standby physical link. The access node may further close the active physical link after completing the service switching (it may be to close the physical port, or simply discard all the service data frames received on the link, and not later) Let me repeat).

A specific implementation block diagram of the protection switching is shown in FIG. 5. In the figure, the service OAM module processes the OAM packet such as the service level continuity fault detection message. When the service level continuity fault is detected, it triggers the protection switching decision module, and activates the standby physical link on the other hand (the main service can be closed at the same time) The physical link), on the other hand, reverses the traffic frame itself to the alternate pseudowire of the alternate LSP path in the alternate physical link.

In the embodiment of the present invention, the maintenance endpoint is configured in the access node and the aggregation node of the service path in the single/multicast service, and the service level continuity fault detection is performed between the maintenance endpoints, thereby eliminating the need for different layers in the path. The MPLS OAM mechanism and the physical layer OAM are deployed between different LSP segments. As long as the service level continuity fault is detected, the access node performs the activation of the standby physical link, whether it is unicast or multicast. The alternate LSP path carried by the standby physical link and the alternate pseudowire carried by the backup LSP path are available, and the service pseudowire carrying the service is switched from the primary pseudowire on the primary LSP path to the alternate pseudowire on the standby LSP path. Therefore, the service protection under the MPLS multi-service access system is realized. This method not only greatly simplifies the service protection process, but also unifies the protection methods of unicast and multicast services. In addition, the method does not need to rely on the intermediate node for service protection, and can protect multiple fault scenarios of multiple services, ensuring the independence of the access device, and reducing the difficulty of service access configuration, and does not need to separately unicast and group. Broadcasting provides different protection mechanisms, making deployment easier. FIG. 6 is a flowchart of another service protection method according to an embodiment of the present invention.

This example uses the VPWS to carry the unicast service as an example. As shown in Figure 7, the access node

There is an active physical link between AN3 and the aggregation node AGS7 AN3-AGS8 and the backup physical link

AN3-AGS9. The primary physical link AN3-AGS8 is in an active state, and its service is carried by the primary pseudowire PW8 on the primary LSP path between the AN3-AGS7, and the standby physical link is used.

AN3-AGS9 is inactive before the service fails, and the service flow cannot be used.

Configure the maintenance endpoint MEP1 on the access node AN3, and configure the maintenance endpoint MEP2 on the aggregation node AGS7. The service level continuity fault detection protocol such as Y.1731 or 802.1ag OAM can be enabled between MEP1 and MEP2.

The service protection method can include the following steps:

Step 601: The access node receives the service level continuity fault detection message sent by the sink node. The MEP2 on the AGS7 sends a service level continuity fault detection message (CCM) to the MEP1 (ie, the downlink direction) on the AN3. The detection message is transmitted through the pseudowire PW8. The source address of the message is the unicast MAC address of the AGS7, and the destination address is Unicast MAC address of AN3.

At the same time, AN3 can also send a service-level continuity fault detection message to AGS7 (that is, the uplink direction). The source address of the message is the unicast MAC address of AN 1, and the destination address is the unicast MAC address of AGS7.

Step 602: The access node determines whether a service fault occurs according to the condition of receiving the service OAM message.

If the access node AN3 also sends a CCM message to the aggregation node AGS7, when the uplink direction fails, the AGS7 cannot receive the CCM message sent by the AN3 within a certain period of time. At this time, the AGS7 sends the reverse CCM message to the AN3 through the downlink direction. The failure indication (for example, a specific bit in the CCM message), so if the AN3 receives the CCM with the reverse fault indication sent by the AGS7 within the preset time, it indicates that there is a fault in the uplink direction of the service. If AN3 receives the CCM message sent by AGS7 within the preset time, and there is no reverse fault indication, it indicates that the service is running normally, that is, it runs normally in the uplink and downlink directions of the service path. If AN3 does not receive any CCM message sent by AGS7 within the preset time (for example, 3.5 CCM cycles), it indicates that the service must have a fault in the downlink direction (it is possible that there is also a fault in the uplink direction).

This service level continuity fault detection is different from the fault detection at the path layer, and it does not need to be different in the path. Different fault detection mechanisms are deployed between the layers to detect the service running between the access node and the aggregation node. The service-level continuity fault detection message is transmitted on the pseudo-line of the service path. As long as a service fault occurs, AN3 It can be perceived that, in turn, it can be determined that the physical link or the LSP path or the intermediate node carrying the service is faulty.

Step 603: When the access node AN3 detects a service level continuity fault, activate the standby physical link of the active physical link that carries the service, and close the primary physical link.

The activation process is similar to the step 402 in the foregoing embodiment, and details are not described herein again. After the standby link AN3-AGS9 is activated, the alternate LSP path on the standby link is available. Since the physical layer link between the AGS9 and the aggregation node AGS7 is always active, the LSP path carried by the LSP is always available. .

After the access node activates the standby physical link AN3-AGS9, the faulty service can be switched to the standby pseudo-line carried by the standby LSP in the standby physical link, and the following steps can be performed to carry the faulty service. All services on the primary physical link are switched to their alternate pseudowires.

Step 604: The access node establishes a service index table of all services on the active physical link.

In this embodiment, in order to facilitate the handover of all services on the active physical link, the access node AN3 may first establish a service index table of all services on the active physical link, as shown in the following table, so as to facilitate subsequent handover. All services can be switched to its alternate pseudowires one by one according to the service index table.

The service index table can take the form: (i, Service(i), PW(i) ), where i is an index value of a positive integer, and Service(i) and PW(i) represent the i-th service and its Corresponding alternate pseudowire, or (i, Service(i), PW main (i), PW standby (i)), where PW main (i) and PW standby (i) respectively represent the main corresponding to the i-th service Use pseudowires and alternate pseudowires. Other possible ways, such as using a linked list, hash, or other data structure to represent the business index table are possible, and are not enumerated here.

The step of establishing the service index table may be established by management pre-configuration, or dynamically established when a new service is accessed, or established when an alternate path is established, which is not limited herein.

Step 605: Switch each service on the active physical link to the corresponding standby pseudowire by using the primary pseudowire that carries each service according to the service index table.

After the standby physical link is activated, all the backup pseudowires carried on the standby physical link are available. In this step, in addition to switching the faulty service from the primary pseudowire carrying the service to its corresponding alternate pseudowire, other services on the primary physical link may be carried by the service. The main pseudowire switches to the corresponding alternate pseudowire. Specifically, the primary pseudowire on the primary physical link may be switched to its alternate pseudowire one by one according to the services listed in the service index table.

If steps 604 and 605 are not performed, and only the failed service is switched from the primary pseudowire to the alternate pseudowire, since the primary physical link has failed, other services on the primary physical link are also running. A fault occurs. Therefore, after the steps of the above steps 601 to 603 are repeated for the corresponding service, the same will be switched to the alternate pseudo line corresponding to the service.

In this embodiment, as shown in FIG. 7, since the unicast service is a homogenous and homogenous scenario, that is, after the access node AN3 is accessed through the primary physical link and the standby physical link, the service is finally exported in the MPLS network. If the same convergence node (for example, AGS7) is used, the primary pseudowire PW8 and the alternate pseudowire PW9 are actually the same pseudowire. The faulty service switching process is specifically as follows: The primary pseudowire carrying the service is used by the primary physics. The primary LSP on the link is switched to the backup LSP on the standby physical link. As shown in Figure 8, the service OAM module processes OAM packets such as service level continuity fault detection messages. When a service continuity fault is detected, It triggers the protection switching decision module, which activates the standby physical link/disables the active physical link on the one hand, and directly switches the service pseudowire to the standby LSP path in the standby physical link. An optional physical link down mode is to discard all service data frames received on the link.

When the unicast service is a homogenous different sink, that is, the primary physical link and the backup physical link are both originated from the access node AN3, but the service is in the MPLS network, the final exit is a different aggregation node, and the primary pseudowire PW8 Different from the alternate pseudowire PW9, the specific handover process of each service is also shown in FIG. 5, which is the same as the handover process in the foregoing step 403.

In the embodiment of the present invention, in the single/multicast service, the maintenance node is configured at the access node and the aggregation node, and the service level continuity fault detection is performed between the maintenance endpoints, thereby eliminating the need to deploy corresponding MPLS OAM between different path layers. The physical layer OAM, as long as the corresponding service fault is detected, the access node actively activates the standby physical link, and switches all services from its primary pseudowire to its corresponding alternate pseudowire, thereby implementing MPLS multi-service access. Business protection under the system. This method not only greatly simplifies the service protection process, but also unifies the protection methods of unicast and multicast services.

FIG. 9 is a flowchart of another service protection method according to an embodiment of the present invention. This example uses VPLS to carry multicast services as an example. As shown in FIG. 10, a physical layer link AN4-AGS11-AGS10 and a physical layer link AN4-AGS 12-AGS 10 exist between the access node AN4 and the sink node AGS10. The active physical link AN4-AGS11 is in the active state, and the service is carried by the pseudowires PW41 and PW10 carried by the multiple LSPs between the AN4 and the AGS10. The standby physical link AN4-AGS12 is in the service before the service fails. Inactive, not available.

Configure the maintenance endpoint MEP3 on the access node AN4 and configure the maintenance endpoint MEP4 on the aggregation node AGS10. The continuity fault detection protocol such as Y.1731 or 802.1ag OAM can be enabled between MEP3 and MEP4.

The service protection method can include the following steps:

Step 901: The aggregation node sends a service level continuity fault detection message to the access node.

The MEP4 on the AGS 10 transmits a service level continuity fault detection message to the MEP3 on the AN4, and the detection message is transmitted in the pseudowires PW41 and PW10. The source address of the message is the unicast MAC address of AGS10, and the destination address is the multicast MAC address.

Step 902: The access node determines, according to the continuity fault detection, whether a service level continuity detection fault occurs.

For example, if the access node MEP3 does not receive the CCM message sent by the MEP4 within 3.5 cycles, it determines that there is a service failure, and further determines that the primary physical link is faulty or other physical link on the primary LSP path or There is a fault in the intermediate node, etc.

Step 903: When the access node AN4 detects a service failure, activate the standby physical link of the active physical link that carries the service, and close the active physical link.

AN4 activates the alternate physical link AN4-AGS12. Since the data paths between AGS12 and AGS10 are always open (they are in the same VPLS network), the alternate LSP path and the alternate pseudowire are available.

An optional physical link closure mode is to discard all service data frames received on the link.

The steps 902 and 903 are similar to the steps 602 and 603 in the foregoing embodiment, and are not described herein again. Step 904: The access node sends an IGMP group (Internet Group Management Protocol) request on the standby physical link by using an alternate pseudowire. After the access node activates the standby physical link, the AGS 10 can further send a MAC flush message to each VPLS endpoint according to the existing mechanism of the VPLS, so that each endpoint re-executes the MAC address learning and accelerates the convergence process of the VPLS service.

For multicast services, AN4 can also choose to perform the IGMP proxy function. On the standby physical link, the proxy client actively sends IGMP group requests, and establishes and maintains multicast group memberships with the directly adjacent multicast routers. The broadcast service is available on the alternate physical link. The IGMP message may be sent by using an alternate pseudowire carried on the standby physical link corresponding to the multicast service. In another embodiment, if there are no alternate pseudowires, the IGMP message can be sent directly on the alternate physical link.

Step 905: Establish a service index table, and switch all services on the active physical link to the corresponding standby pseudowires by using the primary pseudowires respectively carrying the services according to the service index table.

The access node establishes a service index table of all services on the primary physical link, and switches all services on the primary physical link to the corresponding alternate pseudowire according to the service index table. At this time, since the multicast service is transmitted to the AN4 through the pseudowire PW42 on the alternate LSP path in the alternate physical link between the AN4-AGS12, the AN4 can receive the downlink multicast service sent by the AGS10 on the alternate pseudowire. The switching process of each service is also shown in FIG. 5, which is the same as the switching process in the foregoing step 403.

In the embodiment of the present invention, in the multicast service, the maintenance node is configured at the access node and the aggregation node, and the service level continuity fault detection is performed between the maintenance endpoints, thereby eliminating the need to deploy between different path layers.

MPLS OAM and physical layer OAM, as long as a service failure is detected, the access node performs activation of the standby physical link, and all services are switched from their primary pseudowires to corresponding alternate pseudowires, thereby implementing MPLS multi-service. Service protection under the access system. This method not only greatly simplifies the service protection process, but also unifies the protection methods of unicast and multicast services.

In the above embodiment, the BRAS (Broadband Remote Access Server) is a convergence device AGS7/AGS10 that is connected to the upstream edge of the MPLS network at a single point. In another embodiment of the present invention, the BRAS can also be dual-homed to two different aggregation devices, as shown in the figure.

11 is shown.

After the access node switches the service to the backup pseudowire, the service level continuity fault detection may also be performed on the switched service path. If the service fails, the method may continue to perform another standby pseudowire according to the foregoing embodiment. Business switching. The above is a description of an embodiment of the method of the present invention, and an apparatus and system for implementing the above method will be described below.

FIG. 12 is a schematic structural diagram of an access device according to an embodiment of the present invention.

The access device is applied to an MPLS multi-service access system, where the access device is configured with a maintenance endpoint that can perform service level continuity fault detection, and the access device can include:

The detecting unit 1201 is configured to run a service level continuity fault detection on a service path between the node and the sink node. A maintenance endpoint that performs service level continuity fault detection is also configured on the aggregation node.

The activation unit 1202 is configured to activate the standby physical link of the active physical link when the continuity fault of the service level is detected, so that the backup LSP path and the backup pseudowire on the standby physical link are available.

The switching unit 1203 is configured to switch the faulty service from the primary pseudowire of the bearer service to the standby pseudowire, where the primary pseudowire is a pseudowire carried by the primary LSP path on the primary physical link, and the standby The pseudowire is a pseudowire carried by the alternate LSP path on the alternate physical link.

The detecting unit 1201 sends a service level continuity fault detection message between the maintenance endpoint of the access node and the maintenance endpoint of the sink node, and verifies whether there is a service fault according to whether the CCM message of the other party is received within a predetermined time, and the It can be determined whether the bearer between the access node and the sink node is faulty. When it is detected that the service fails, the activation unit 1202 abandons the use of the active physical link, activates the standby physical link, and after the standby physical link is activated, the switching unit 1203 carries the primary pseudowire on the primary physical link. The service is switched to the alternate pseudowire on the standby physical link, and then the service is carried by the alternate pseudowire.

The embodiment of the present invention implements the service protection in the MPLS multi-service access system by using the above-mentioned unit, greatly simplifies the service protection process, and unifies the protection methods of the unicast and multicast services, thereby ensuring the independence of the access device and reducing the independence. The difficulty of service access does not require different protection mechanisms for unicast and multicast, which makes deployment easier.

FIG. 13 is a schematic structural diagram of another access device according to an embodiment of the present invention.

The access device may include a detecting unit 1301, an activating unit 1302, and a switching unit 1303. The detecting unit 1301 and the activating unit 1302 are respectively similar to the detecting unit 1201 and the activating unit 1202 in the foregoing embodiment, and details are not described herein again.

In this embodiment, the detecting unit 1301 is specifically configured to enable service level continuity fault detection of the Y.1731 or 802. lag OAM between the access device and the sink node. The switching unit 1303 may further include:

A sub-unit 13031 is created, which is used to establish a service index table for all services on the active physical link. The switching subunit 13032 is configured to switch, according to the service index table, each service on the active physical link by the primary pseudowire that separately carries each service to the corresponding standby pseudowire.

When the service is a unicast service and the same as the same, the primary pseudowire and the alternate pseudowire are the same pseudowire. Therefore, the switching unit may be specifically configured to use the primary pseudowire carrying the service by the primary physical chain. The primary LSP on the road directly switches to the backup LSP on the standby physical link.

In another embodiment, for the multicast service, the access device may further include a request sending unit, configured to pass the failed pseudo-line before switching the failed service to the backup pseudowire by using the primary pseudowire carrying the service. An Internet Group Management Protocol IGMP group request message is sent on the alternate physical link.

In another embodiment, the access device may further include a shutdown unit, configured to close the service after the faulty service is switched by the primary pseudowire carrying the service to the backup pseudowire. Primary physical link.

FIG. 14 is a schematic structural diagram of an MPLS multi-service access system according to an embodiment of the present invention. The system can include an access device 1401 and a convergence device 1402.

The access device 1401 is configured to run a service level continuity fault detection on the service path between the aggregation device 1402, where the access device is configured with a maintenance endpoint that can perform the service level continuity fault detection; When a continuity fault of the service level is detected, the standby physical link of the active physical link is activated, so that the backup LSP path and the backup pseudowire on the standby physical link are available; The primary pseudowire of the service is switched to the standby pseudowire, where the primary pseudowire is a pseudowire carried by the primary LSP path on the active physical link, and the standby pseudowire is A pseudowire carried by a backup LSP path on the standby physical link.

The aggregation device 1402 is configured to perform service level continuity fault detection on the service path with the access device 1401, and the aggregation device 1402 is configured with a maintenance endpoint that can perform service level continuity fault detection. The system greatly simplifies the service protection process, unifies the protection methods of unicast and multicast services, ensures the independence of access devices, and reduces the difficulty of service access. It does not need to provide different unicast and multicast respectively. Protection mechanisms make deployment easier.

The embodiments of the present invention described above are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the invention are intended to be included within the scope of the appended claims.

Claims

Rights request

A service protection method, characterized in that:

The method according to claim 1, wherein the switching the faulty service from the primary pseudowire carrying the service to the standby pseudowire includes:

Establishing, by the access node, a service index table of all services on the active physical link; and switching, according to the service index table, each service on the active physical link by using a primary pseudowire that carries each service separately Go to the corresponding alternate pseudowire.

The method according to claim 1 or 2, wherein when the service is a multicast service, the service that is to be failed is switched to the standby by the primary pseudowire carrying the service. Before the pseudowire, it also includes:

The access node sends an Internet Group Management Protocol IGMP request message on the standby physical link through the alternate pseudowire.

The method according to claim 1, wherein when the service is a unicast service and the same as the same, the primary pseudowire and the alternate pseudowire are the same pseudowire, and the method will appear. The faulty service is switched to the standby pseudowire by the primary pseudowire carrying the service, specifically:

The primary pseudowire carrying the service is switched from the primary LSP path on the primary physical link to the alternate LSP path on the alternate physical link.

The method according to any one of claims 1 to 4, further comprising: after the failing service is switched by the primary pseudowire carrying the service to the standby pseudowire , the primary physical link is closed.

6. An access device, comprising:

The access device according to claim 6, wherein the switching unit comprises: a creating subunit, configured to establish a service index table of all services on the active physical link; And switching, according to the service index table, each service on the primary physical link to a corresponding standby pseudowire by using a primary pseudowire that carries each service.

The access device according to claim 6 or 7, further comprising: a request sending unit, configured to: when the service is a multicast service, the service that is to be faulty is carried by the service Before the primary pseudowire switches to the standby pseudowire, the IGMP group request message is sent on the standby physical link by using the alternate pseudowire.

9. The access device of claim 6, wherein

The switching unit is specifically configured to: when the service is a unicast service and the same as the same, the primary pseudowire and the standby pseudowire are the same pseudowire, and the primary pseudowire that carries the service is Switching from the primary LSP path on the primary physical link to the alternate LSP path on the alternate physical link.

The access device according to any one of claims 6 to 9, further comprising:

And a shutting down unit, configured to close the active physical link after the faulty service is switched by the primary pseudowire carrying the service to the standby pseudowire.