WO2006089490A1

WO2006089490A1 - A method for implementing fec bfd

Info

Publication number: WO2006089490A1
Application number: PCT/CN2006/000272
Authority: WO
Inventors: Suping Zhai
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2005-02-28
Filing date: 2006-02-24
Publication date: 2006-08-31
Also published as: CN100525244C; CN1829191A

Abstract

A method for implementing FEC BFD comprises: firstly, setting FEC filter information in BFD control message; then, implementing BFD mechanism in MPLS networks using BFD control message comprising FEC filter information. Therefore, the present invention enhances BFD detection mechanism to implement the function of LSP FEC misbranching detection and provides a feasible technical solution for LSP FEC misbranching detection of MPLS networks. The present invention also records the nested relationship of LSP when establishing LSP to implement the failure announce mechanism of the down stream.

Description

TECHNICAL FIELD The present invention relates to the field of network communication technologies, and in particular, to a method for implementing an equivalent forwarding type bidirectional forwarding detection in an MPLS network.

Background technique

OAM (Operation and Maintenance) technology is a technology that provides operational and maintainable functions. It can automatically detect various failures in the network and then take corresponding rescue measures to avoid the expansion of faults or eliminate them as much as possible. The negative impact of the failure.

The MPLS OAM technology standard of the ITU-T (International Telecommunication Union-Telecommunication Standardization Division) is a detection mechanism for MPLS (Multiprotocol Label Switching) LSP (Label Switching Path), that is, sending a Hello (Handshake) message between two points. A way to detect a link. In Y.1713, FEC-CV messages are defined to detect the connectivity of FEC (equivalent forwarding class).

The format of the payload of the FEC-CV message is as follows:

The LSR ID (label switching router identifier) and the LSP ID (label switching path identifier) form a TTSI (the endpoint source identifier of the LSP connection);

Assume that Node A and Node B are the endpoints of the LSP connection, and the LSP is established through the LDP (Label Distribution Protocol) (RFC3036). The process of detecting the FEC through the FEC-CV packet includes the following steps:

Step 1: Establish an LSP channel between Node A and Node B. In the process of establishing an LSP, an FEC Filter (FEC filter) entry is established at the A node by using the FEC Element (FEC element) in the LDP protocol, and the entry includes the TTSI (the endpoint source identifier of the LSP connection) and the FEC Element. For searching and detecting FEC, the TTSI includes an LSR ID (Label Switching Router Identity) and an LSP ID (Label Switching Path Identity).

Step 2: Send the FEC-CV packet from the node A to the node B with the configured T ^FEC - ^CV as the time interval. The packet carries the FEC Filter field, which carries the FEC Filter information, that is, the FEC. The element information is used to detect the correspondence between the FECs, that is, the FEC detection function is implemented; the FEC Filter information is calculated by the FEC Element information by the method specified by Y.1713; Step 3: The FEC- received at the Node B After the CV packet, the corresponding FEC is found on the Node B by using the TTSI information carried in the packet. Then, the found FEC Element is divided into three filter offsets (FEC filters) according to the method defined in Y.1713, and the FEC Filter of the Node B is calculated by using the three filter offsets and the FEC Element;

Step 4: Compare the FEC Filter (FEC Filter of Node A) and the calculated FEC Filter (FEC Filter of Node B) in the received packet by using the corresponding matching algorithm to determine whether the dFEC- Mismatch FEC corresponding occurs. Wrong fault) / dFEC - Mismerge ( FEC cross fault fault);

(1) For the exact matching algorithm, directly compare the FEC Filter in the received message with the calculated FEC Filter. If they are equal, the LSP FEC (label switching path forwarding equivalence class) is correct, otherwise dFEC_Mismatch/dFEC is generated. — Mismerge;

(2) For the subset matching algorithm, use the FEC Filter in the received message and the calculated FEC Filter to reverse the summation. If it is 0, it indicates that the LSP FEC is correct, otherwise dFEC-Mismatch/dFEC-Mismerge is generated.

It can be seen that the above disadvantages of the FEC detection mechanism defined by 丫.1713 are as follows: Once the FEC detection mechanism is activated, the OAM mechanism will continue to operate unless the link fails, and the configuration cannot be flexibly performed.

Currently, in order to quickly switch to other links in the event of a failure, it is necessary to quickly find the fault. Some hardware, such as SDH, can provide BFD (Bidirectional Forwarding Detection), but this feature is not available for many hardware or software, such as Ethernet. There are also some that cannot implement path detection, such as forwarding engines or interfaces, and cannot achieve end-to-end detection. Currently, the network generally uses a slow Hello (handshake) mechanism, especially in a routing protocol. Without hardware help, the detection time is long and cannot be detected for nodes that do not have a routing protocol allowed.

The goal of BFD is to provide a low-overhead, short-detection failure detection mechanism on the path between adjacent forwarding engines. Includes interfaces, data links, and extensions to the detection of the forwarding engine itself. The basic detection mechanism of BFD is shown in Figure 1, where:

The BFD inspection process consists of two optional modes of operation, and an accompanying function, which can be mixed.

The first mode of operation is asynchronous mode: In this mode, the system periodically sends BFD control messages to another system. If another system does not receive a certain number of messages, it determines that the corresponding connection needs to announce an interrupt ( Down).

The second mode is Demand mode: In this mode, it is assumed that each system has an independent way to verify its connection with other systems, so once a BFD connection is established, the system will stop sending BFD control reports. In this case, unless another system explicitly believes that the link needs to be verified, a short sequence of BFD control messages is sent and the protocol is in a quiescent state.

The other is an additional function for the two modes, namely the echo function: When the echo function is activated, the BFD echo packet stream is sent back to the peer forwarding path; if a certain number of packets in the sequence are not received, The connection is declared to be down (down); the echo function can be used in the above two modes. If the detection task is implemented by the echo function, the periodic message can be reduced in the asynchronous mode, and the Hello message can be completely cancelled in the on-demand mode.

The BFD control packet is carried by the UDP (User Datagram Protocol) packet. The UDP destination port of the packet is the source port of the 3784 source port is 49252 ~ 65535. The destination port of the response packet is 3785.

The methods of the BFD detection path are as follows:

Asynchronous mode: BFD packets are sent in both directions between nodes A and B. If a certain number of packets are lost within a certain period of time, the link is considered to be broken.

Echo function: If the node supports, the packet can be sent from the node A. The node B only loops back the packet and sends it back to the node A. The node A detects the packet loss and determines whether the link is available.

On-demand mode: Nodes A and B can also negotiate between links only when needed. Between nodes A and B, the speed at which packets can be sent can be negotiated and the other party can tell each other that they can receive and The speed of the send.

However, because the BFD detection mechanism is for MPLS LSP detection, there is no mechanism for determining how to detect the LSP FEC, and there is no mechanism for suppressing the alarm, that is, for the failure, There is no implementation that provides an alert notification to downstream nodes.

Summary of the invention

In view of the above problems in the prior art, the object of the present invention is to provide a method for implementing bidirectional forwarding detection of equal-cost forwarding, so that the detection of cross-error for LSP FEC can be implemented based on the BFD detection mechanism in the M PLS network.

The object of the invention is achieved by the following technical solutions:

The present invention provides a method for implementing bidirectional forwarding detection of equivalent forwarding type, including:

A. Set the equivalent forwarding class filter information in the BFD control packet in the bidirectional forwarding detection packet;

B. Implementing a BFD mechanism in a multi-protocol label switching MPLS network by using BFD control messages including information of equivalent forwarding class filters.

The equivalent forwarding class filter information includes:

Label Switching Path The endpoint source identifier of the LSP connection and the information of the forwarding class element. In the MPLS network, the node that receives the BFD control packet detects whether the equivalence forwarding class has a cross fault fault according to the equivalent forwarding class filter information carried in the packet. On the node that receives the BFD control packet, if the BFD detection state machine is in the active state (UP), the matching detection is performed according to the equivalent forwarding class filter information in the packet to determine whether the equivalent forwarding class crosses. Wrong fault.

Matching detection is performed by using a matching algorithm or a subset matching algorithm to determine whether an equivalent forwarding class has a cross fault.

The method for implementing the bidirectional forwarding detection of the equivalent forwarding type further includes:

The BFD control packet further includes defect coding information, which is used to identify the cross-error information of the equivalent forwarding class of the local node.

When the LSP is set up, the LSP nesting relationship is recorded. The mapping between the LSP ID and the source IP address and destination IP address is recorded.

C. The correspondence between the LSP identifier and the source IP address and the destination IP address is determined. IP address information of the tour node;

D. Send the BFD control packet carrying the defect coding information to the downstream according to the IP address information of the determined downstream node. During the establishment of the BFD session, record the local discriminator My Discriminator information and

LSP identifier, and the correspondence between the source IP address and the destination IP address;

According to the My Discriminator information and the recorded information in the received BFD control packet, the destination IP address information of the corresponding downstream node is found.

It can be seen from the technical solution provided by the present invention that the BFD mechanism is enhanced, thereby implementing the detection function for the LSP FEC cross error, and providing a feasible technical solution for detecting the FEC error in the MPLS network. . At the same time, when the LSP is established, the LSP nesting relationship is recorded, so that the downstream fault notification mechanism is implemented, that is, the downstream alarm mechanism is implemented.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of a conventional BFD detection state;

2 is a flow chart of the method of the present invention;

3 is a schematic diagram of a format of a BFD control packet in the present invention;

Fig. 4 is a schematic view showing the state of BFD detection in the present invention.

The core of the present invention is to enhance the detection type of BFD, increase the defect type of dFEC-Mismatch, and implement the detection of MPLS FEC cross error by using BFD mechanism. Meanwhile, it is also proposed to use LSP established between LSP establishment. Nested relationship, the downstream notification mechanism for failure.

A specific implementation manner of the method of the present invention is shown in FIG. 2, and specifically includes:

Step 21: In order to implement FEC cross-error detection, it is necessary to expand the existing BFD control message, and add an FEC Fier ter (FEC filter) field for carrying FEC filter information, the FEC. The filter information is calculated by the FEC Element information by the method specified by Y. 1713. The information of the equivalent forwarding class filter includes: an end point source identifier of the label switched path LSP connection and the information of the forwarding class element. That is, the method of the present invention is implemented based on the BFD detection mechanism. In this embodiment, the format of the BFD control packet used to implement BFD detection is as shown in FIG. 3, except for the FEC Filter in the figure. It is inherent in BFD, so only the FEC Filter field is described.

FEC Filter field: used to carry FEC Filter information obtained according to FEC Element information, and the FEC Filter information is used to implement FEC misbranching (FEC cross error or corresponding error) detection; the meaning and function of the field are related to the prior art. The meaning and function of this field in the FEC-CV message are the same.

The FEC Filter information is newly added information based on the existing BFD control packet format, and the specific setting manner may refer to 171713 Misbranching detection for MPLS networks (Y.1713 MPLS network cross error detection).

Step 22: performing FEC cross-error fault detection by using the BFD control packet containing the FEC lter information;

The specific detection method used includes: receiving, by the node of the BFD control message, the FEC filter information according to the F艮 Filter information, and adopting an exact matching algorithm or a subset matching algorithm to perform FEC cross fault detection;

The specific implementation method and the prior art for the detection of the cross-error fault by using the FEC Filter information Therefore, the detailed description of the process will not be repeated here.

In this embodiment, the BFD detection state machine in the node needs to be adaptively improved, as shown in FIG. 4: Since the BFD packet carries the FEC Filter information, the node receives the BFD control. If the state machine of the BFD detection on the node is in the UP state, you need to use the FEC Filter information carried in the packet to perform matching detection to determine whether the FEC has a fault.

Step 23: In the present invention, the defect coding information is added to the BFD control message, that is, dFEC-Mismatch, which is used to carry the FEC fault error information;

In this way, when the FEC has a fault of cross error, the fault information can be carried on the fault information.

The BFD control message is transmitted to other nodes (ie, downstream nodes);

The defect number in the BFD control packet can be defined as 8; of course, other values can also be used. Step 24: In the method of the present invention, in order to cooperate with the BFD detection mechanism to implement the function of the downstream alarm (that is, the cross-error fault information of the FEC is transmitted downstream), in the process of establishing the LSP, it is also required to record the inter-inlay between the LSPs. Set of relationships;

The specific implementation manner for recording the nested relationship between the LSPs may be: Record LSP ID (LSP ID) and <31 (source IP address), DIP (destination IP address)

In the process of establishing a BFD session, the local discriminator My Discriminator information needs to be associated with the above LSP ID, <SIP, DIP>, so that the My Discriminator can be used to index the LSP ID and corresponding thereto. <31, DIP>, so that the address information of the remote node can be determined, thereby determining the address information of the destination node when the alarm information is sent;

The My Discriminator information is an authentication value of the MPLS LSR according to different BFD sessions, and a My Discriminator information corresponds to an FEC connection.

Step 25: When a FEC cross fault fault occurs on the local node, the LDP nested relationship table is searched through the My Discriminator information to obtain the IP address information of the downstream node.

Step 26: On the local node, start the BFD failure notification packet, and carry the corresponding defect coding information in the BFD control packet, and send it to the downstream node.

When the local node has other faults, set Sta to Down in the BFD control packet.

The Diag (diagnostic) field is 5 ( Administratively Down), indicating that the underlying LSP is faulty, the upper layer.

The LSP needs to be forced into the Down state to suppress the generation of alarm packets to avoid alarm storms in the network. The network node does not know the specific faulty network segment or node.

In summary, it can be seen that the present invention solves the shortcomings of the FEC detection mechanism defined by Y.1713, and also maintains the advantages of the BFD detection mechanism, that is, the BFD detection mechanism can implement the FEC detection function. Therefore, the present invention realizes the function of flexibly configuring the detection mode of the FEC detection process, such as flexibly configuring the asynchronous mode, the on-demand mode, and the echo function according to actual needs, and implementing FEC detection based on the corresponding mode or function. In addition, the present invention further expands the downstream alarm function so that various fault information such as detected cross error and connection relay can be transmitted to the downstream node.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, and any technical person skilled in the art can disclose the technical scope disclosed by the present invention. Variations or substitutions that are conceivable within the scope of the invention are intended to be included within the scope of the invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

Rights request

A method for implementing bidirectional forwarding detection of an equivalent forwarding type, characterized in that:

A. Set the equivalent forwarding class filter information in the BFD control packet to the forwarding detection.

The method for implementing the bidirectional forwarding detection of the equivalent forwarding type according to claim 1, wherein the information of the equivalent forwarding class filter includes:

Label Switching Path The endpoint source identifier of the LSP connection and the information of the forwarding class element.

3. The method for implementing equivalent forwarding type bidirectional forwarding detection according to claim 1, wherein:

MPLS network

The node that receives the BFD control packet detects whether the equivalence forwarding class has a cross fault fault according to the equivalent forwarding class filter information carried in the packet.

The method for realizing the equivalent forwarding type bidirectional forwarding detection according to claim 1, 2 or 3, characterized in that:

On the node that receives the BFD control packet, if the BFD detection state machine is in the active state UP, the matching detection is performed according to the equivalent forwarding class filter information in the packet to determine whether the equivalent forwarding class has a cross fault. .

The method for realizing the equivalent forwarding type bidirectional forwarding detection according to claim 1, 2 or 3, characterized in that

The method for implementing the equivalent forwarding type bidirectional forwarding detection according to claim 4, wherein the matching algorithm or the subset matching algorithm is used for matching detection.

The method for implementing the equivalent forwarding type forwarding forwarding detection according to claim 5, wherein the method further comprises:

8. The method for implementing equivalent forwarding type bidirectional forwarding detection according to claim 7, wherein The levy also includes:

C. determining the IP address information of the downstream node according to the correspondence between the LSP identifier and the source IP address and the destination IP address;

D. Send the BFD control packet carrying the defect coding information to the downstream according to the IP address information of the determined downstream node.

9. The method for implementing equivalent forwarding type bidirectional forwarding detection according to claim 8, wherein:

During the establishment of the BFD session, the local discriminator My Discriminator information and the LSP identifier, and the mapping between the source IP address and the destination IP address are recorded.

10. The method for implementing equivalent forwarding type bidirectional forwarding detection according to claim 8, wherein:

According to the My Discriminator information and the recorded information in the received BFD control packet, the destination ip address information of the corresponding downstream node is searched.