Bidirectional Fault Detection in LSP Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current troubleshooting tools for bidirectional interactive services, such as LSP Ping and LSP Trace, are limited to unidirectional handling modes, leading to increased signaling and processing overhead, network load, and operational complexity, making it difficult to efficiently detect, locate, and handle faults in both forward and backward paths simultaneously.
Innovation Solution
A method and apparatus for bidirectional troubleshooting that involves sending a forward detection message carrying information about both forward and backward paths, allowing simultaneous detection and verification of path consistency between the data and control planes, enabling fault detection and handling in both directions with reduced signaling overhead and simplified operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If unidirectional troubleshooting tools (LSP Ping, LSP Trace) are used to detect faults in both forward and backward paths, then fault detection capability is provided, but signaling overhead and processing overhead increase significantly
Solution Approach 1:
The patent merges forward path detection and backward path detection into a single bidirectional detection message. The detection initiation node sends one message that carries both forward path information (to be detected by the detection target node) and backward path information (to be detected by intermediate nodes), eliminating the need for separate unidirectional detection messages in each direction.
Solution Approach 2:
The bidirectional detection message serves multiple functions simultaneously: it detects the forward path from initiation node to target node, detects the backward path from target node to initiation node through intermediate nodes, and enables fault location in both directions. This multi-functional message reduces the total number of detection messages required.
2Reliability
If unidirectional troubleshooting tools are operated on both sides independently to detect faults in two directions, then complete bidirectional fault detection is achieved, but operational complexity increases
Solution Approach 1:
Instead of having both endpoints independently initiate unidirectional detection messages, the patent inverts the approach by having the detection initiation node send a single bidirectional detection message that automatically triggers detection in both directions. The intermediate nodes on the backward path actively participate in detecting the backward path, reversing the traditional passive role of intermediate nodes.
Solution Approach 2:
The bidirectional detection message enables the network itself to perform self-diagnosis by having intermediate nodes automatically detect the backward path using information contained in the detection message. This self-service mechanism eliminates the need for manual configuration and operation at both endpoints, reducing operational complexity.
3Reliability
If separate unidirectional detection messages are sent in both directions, then fault detection is performed, but the number of detection messages and network load increase
Solution Approach 1:
The patent combines forward path detection requirements and backward path detection requirements into a single bidirectional detection message. This message contains both forward path information (destination address, forward path identifier) and backward path information (intermediate node addresses, backward path identifier), allowing one message to perform the work of multiple separate messages.
4Productivity
If bidirectional LSPs with same route and LSP ID are used, then resource efficiency is improved, but fault detection and location become more difficult
Solution Approach 1:
The patent segments the detection functionality by including separate forward path information and backward path information within the same detection message. Even though the forward LSP and backward LSP share the same route and LSP ID, the detection message treats them as distinct entities with separate identifiers, allowing independent fault detection and location for each direction while maintaining resource efficiency.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
A troubleshooting method and a troubleshooting apparatus are disclosed. The troubleshooting method includes: sending a forward detection message to at least one node on a path to be detected, where the forward detection message carries information about a forward path to be detected and information about a backward path to be detected; and detecting faults according to a backward detection message returned by the at least one node. With the troubleshooting method and the troubleshooting apparatus disclosed herein, troubleshooting operations such as fault detection and fault location can be performed for the forward path and the backward path that require bidirectional path detection.