Network Diagnosis Processing via Multi-Layer Tunnel Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network diagnosis methods face inefficiencies and inaccuracies due to the need for multiple independent Operation Administration and Maintenance (OAM) procedures across various network layers and management domains, limiting the ability to effectively process and analyze diagnosis reports.
Innovation Solution
A method and apparatus for network diagnosis processing that involves sending diagnosis messages across multiple network layers, carrying characteristic information to facilitate cross-layer and cross-domain analysis, allowing for the association of diagnosis results and improving efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple independent OAM procedures are used across various network layers and management domains, then each layer can perform local diagnosis, but the efficiency and accuracy of processing and analyzing diagnosis reports deteriorates
Solution Approach 1:
The patent merges multiple independent OAM procedures into a unified diagnosis mechanism by introducing a diagnosis message format that can be processed across multiple network layers (VXLAN tunnel layer, Ethernet layer, IP layer) simultaneously. The instruction center coordinates these procedures to share diagnosis results, eliminating redundant work and improving overall processing efficiency while maintaining comprehensive coverage.
Solution Approach 2:
The diagnosis message is designed with universal characteristics that allow it to be processed and recognized across different network layers and management domains. The message includes universal identification fields and can be handled by multiple types of network devices (routers, switches, tunnels), making the diagnosis system multi-functional and applicable throughout the entire network stack.
2Reliability
If multiple independent OAM procedures are used across various network layers and management domains, then each layer can perform local diagnosis, but the accuracy of analyzing and associating diagnosis reports deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where diagnosis results from different network layers are fed back to the instruction center, which then correlates these results using characteristic information (such as flow identifiers, tunnel IDs, and session parameters). This feedback loop enables accurate association of diagnosis reports across layers by continuously comparing and matching results based on shared characteristics.
Solution Approach 2:
The instruction center acts as an intermediary that receives diagnosis reports from multiple network layers and management domains, then mediates the correlation process by using characteristic information as a key. This intermediary function enables accurate association of diagnosis reports that would otherwise remain isolated, improving measurement precision through centralized coordination.
3Ease of operation
If OAM is terminated at a peer end of a link, then link-layer diagnosis is simple, but the ability to diagnose end-to-end communication failures deteriorates
Solution Approach 1:
The patent applies the nesting principle by embedding OAM diagnosis messages within multiple network protocol layers. The diagnosis message is nested in the VXLAN tunnel header, which itself is nested within the Ethernet frame, which is nested within the IP packet. This nested structure allows the same diagnosis mechanism to operate at multiple layers simultaneously, enabling end-to-end diagnosis while maintaining the simplicity of link-layer OAM implementation.
Solution Approach 2:
The patent extends OAM from a single link-layer dimension to multiple network layer dimensions by introducing diagnosis procedures at the VXLAN tunnel layer, Ethernet layer, and IP layer. This dimensional expansion allows the system to diagnose failures across the entire communication path rather than being limited to a single link, significantly improving end-to-end diagnosis capability.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
The present invention provides a method and an apparatus for network diagnosis processing. The method for network diagnosis processing of the present invention includes: receiving, by a first network tunnel source end, at a first network layer, a network diagnosis instruction sent by an instruction center; sending, by the first network tunnel source end, at the first network layer, a diagnosis message to a first network layer of a first network tunnel peer end and a second network layer of the first network tunnel source end, where the diagnosis message carries characteristic information, so that the first network tunnel peer end reports, at the first network layer, a diagnosis result including the characteristic information to the instruction center; and sending, by the first network tunnel source end, at the second network layer, the diagnosis message to a second network layer of the first network tunnel peer end, so that the first network tunnel peer end reports, at the second network layer, a diagnosis result including the characteristic information to the instruction center. Embodiments of the present invention resolve problems of low efficiency and low accuracy in a process of processing, analyzing and associating diagnosis results among various network layers or various network management domains.