Overlay Network Loop Origination Detection Through Recursive Probing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network management systems struggle to efficiently identify and correct loop-causing switches in overlay networks, particularly when loops are initiated by misconfigured forwarding rules or end devices outside their control.
Innovation Solution
A network management system recursively probes switches along a loop path to determine ports observing a MAC address, identifies switches that have incorrectly forwarded packets, and marks them as loop origination points (LOPs) by analyzing forwarding rule violations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If network management systems use traditional loop detection methods, then loop detection capability is provided, but the ability to efficiently identify and correct loop-causing switches is insufficient
Solution Approach 1:
The system implements a feedback mechanism where loop error information is propagated back through the overlay network path. When a switch detects a loop, it sends loop error information to upstream switches, enabling them to identify themselves as potential loop-origination points. This feedback loop allows precise identification of loop-causing switches without requiring complex manual investigation.
Solution Approach 2:
The patent introduces loop error information as an intermediary carrier that transmits loop detection data between switches. This intermediary mechanism enables switches to communicate loop-related information without direct intervention, allowing efficient identification of loop-origination points through automated information exchange rather than complex detection algorithms.
2Measurement precision
If recursive probing is performed on all switches in the loop path, then loop-origination points can be identified, but network overhead and processing time increase
Solution Approach 1:
The system extracts only the essential loop error information (loop detection status and path identification) from the complex probing process. By focusing on extracting and propagating this specific information rather than performing exhaustive probing of all switches, the system achieves accurate loop-origination point identification with reduced processing time and network overhead.
Solution Approach 2:
Switches perform preliminary loop detection and generate loop error information before full recursive probing is initiated. This preliminary action allows the system to quickly identify potential loop paths and focus subsequent probing efforts only on relevant switches, significantly reducing overall processing time while maintaining identification accuracy.
3Reliability
If loop detection is implemented in overlay networks with multiple fabrics, then comprehensive loop detection is achieved, but complexity of identifying loop-causing switches increases
Solution Approach 1:
The system segments the overlay network into multiple fabrics and implements independent loop detection within each fabric. Each fabric processes loop error information separately, which simplifies the identification process by breaking down the complex multi-fabric problem into manageable segments. This segmentation maintains comprehensive detection coverage while reducing overall system complexity.
Solution Approach 2:
The patent adds a new dimension to loop detection by introducing fabric-specific path identification alongside traditional loop detection. This dimensional enhancement allows the system to track loop paths across multiple fabrics using structured path information, making it easier to identify loop-origination points in complex multi-fabric environments without increasing fundamental detection complexity.
Data Source
AI summary
A network management system (NMS) for provisioning and managing an overlay network is provided. During operation, the NMS can determine that a loop associated with a media access control (MAC) address is detected in the network. The NMS can probe a first switch in the loop by instructing the first switch to observe the MAC address for a predetermined period. The NMS can receive a first message indicating a first set of ports of the first switch observing the MAC address. The NMS can probe an upstream switch reachable via each of the first set of ports by instructing the upstream switch to observe the MAC address for the predetermined period. Here, observing the MAC address at an upstream port of the upstream switch causes further upstream probing. Based on probing the switches in the loop, the NMS can determine one or more loop origination points (LOPs) for the loop.