Segment Routing Path Fault Detection Using Strict Path Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In segment routing (SR) scenarios, the seamless bidirectional forwarding detection (SBFD) control packet may bypass faulty paths, leading to incorrect path status detection and bandwidth waste due to traffic being forwarded through bypass paths.
Innovation Solution
A network device uses a segment list with loose path identifiers to forward a packet and detect the status of a target path, incorporating mechanisms like hop count limits and interface consistency checks to prevent bypass forwarding and ensure accurate path detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If SBFD control packet uses loose path segment identifiers to allow flexible forwarding, then forwarding flexibility is improved, but path detection accuracy deteriorates because the packet may bypass faulty paths
Solution Approach 1:
The patent segments the path detection process into two independent phases: first determining the strict path based on segment list, then detecting status along this strict path. This segmentation resolves the contradiction by separating path determination (which allows flexibility) from path detection (which requires precision through strict path following).
Solution Approach 2:
Instead of using loose path identifiers for detection as in conventional SBFD, the patent inverts the approach by using strict path identifiers derived from the segment list to ensure accurate detection. The strict path is constructed by extracting specific segment identifiers that define the exact routing path, ensuring detection accuracy while maintaining the flexibility of loose path forwarding for data packets.
2Reliability
If SBFD control packet allows bypassing faulty paths through loose path identifiers, then packet delivery success is improved, but bandwidth waste increases due to traffic being forwarded through bypass paths
Solution Approach 1:
The patent implements feedback mechanisms where the head node receives detection results from transit nodes and uses this information to determine the actual strict path. This feedback loop ensures that only valid paths are selected for data forwarding, preventing bandwidth waste while maintaining delivery success. The strict path determination based on feedback from multiple transit nodes ensures optimal path selection.
Solution Approach 2:
The patent performs preliminary path status detection before actual data forwarding occurs. By first determining the strict path and verifying its status using control packets, the system prevents bandwidth waste from the outset. The strict path is pre-calculated and validated, ensuring that only efficient paths are used for data transmission and avoiding unnecessary bypass routing.
3Adaptability or versatility
If the segment list includes multiple segment identifiers for loose path, then path selection flexibility is improved, but detection complexity increases due to multiple possible paths
Solution Approach 1:
The patent segments the complex detection task into manageable parts by processing each segment identifier sequentially. The strict path is constructed by extracting segment identifiers one by one from the segment list, and detection is performed along this segmented path. This segmentation reduces detection complexity while preserving the flexibility of having multiple segment identifiers in the original segment list.
Solution Approach 2:
The patent applies local quality by treating each segment identifier differently based on its role in the path. Not all segment identifiers are treated equally - some are used to construct the strict path while others provide alternative routes. This differentiated treatment simplifies detection by focusing on the critical segment identifiers that define the primary path, while maintaining flexibility through secondary identifiers.
Data Source
AI summary
A path fault detection method, including obtaining, by a network device, a first packet, where the first packet has a segment list, where the segment list has a plurality of sequentially arranged segment identifiers (SIDs), where the segment list identifies a forwarding path of the first packet, and where the segment list has a segment identifier of a first node and a segment identifier of a second node, forwarding, by the network device, the first packet based on the segment list, and detecting, by the network device, a status of a target path based on the first packet, where the target path is a shortest path between the first node and the second node, and where a segment identifier that is in the segment list and that identifies the target path indicates a loose path.