Inter-Chassis Link Failure Management via Backup Control Channel
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional solutions to Inter-Chassis Link (ICL) failures in information handling systems, such as those using Virtual Link Trunking (VLT), result in reduced availability and bandwidth of the VLT fabric, leading to traffic loss and blockage of 'east-west' traffic, as they typically disable VLT ports on secondary peer devices when an ICL fails.
Innovation Solution
An ICL failure management system that determines the unavailability of the primary control channel and adds a first and second port to create a primary backup control channel within the control network, enabling the transmission of control information through these ports to maintain network functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If VLT ports on secondary peer device are disabled when ICL fails, then control information exchange is prevented and temporary loops are avoided, but network availability is reduced by 50% and bandwidth is reduced leading to traffic loss
Solution Approach 1:
The system segments the control information exchange path by separating the data plane (VLT ports) from the control plane (ICL). When ICL fails, only control plane communication is affected while data plane remains operational. This allows continued east-west traffic flow and maintains network availability while still providing control information exchange through alternative paths.
Solution Approach 2:
The patent introduces an intermediary mechanism where control information can be exchanged through the data plane paths even when the direct ICL control path fails. This intermediary path allows VLT peer devices to continue exchanging control information through the network fabric rather than requiring direct ICL connectivity, thereby maintaining both reliability and productivity.
2Stability of the object's composition
If VLT ports on secondary peer device are disabled when ICL fails, then incorrect hashing and temporary loops are prevented, but east-west traffic is blocked and overall bandwidth is reduced
Solution Approach 1:
The system dynamically adjusts the operational state of VLT ports based on ICL status. Rather than statically disabling ports when ICL fails, the system maintains port availability and uses dynamic routing/control mechanisms to handle the failure condition. This allows the network to adapt to failures while maintaining traffic flow and avoiding the harmful effects of static port disabling.
Solution Approach 2:
The patent changes the operational parameters of VLT ports from a binary enabled/disabled state to a more nuanced state where ports remain enabled but control information exchange uses alternative paths. This parameter change allows the system to maintain traffic flow while still providing protection against incorrect hashing and temporary loops through enhanced control plane logic.
3Reliability
If conventional ICL failure management is used, then control information exchange is maintained through ICL, but VLT fabric availability is reduced and bandwidth is reduced causing traffic loss
Solution Approach 1:
The system makes the data plane paths multi-functional by using them for both data transmission and control information exchange. When ICL fails, the same physical paths that carry data traffic also carry control information, eliminating the need for separate dedicated control paths. This universality allows full bandwidth utilization for both data and control traffic, preventing the bandwidth reduction caused by conventional approaches.
Data Source
AI summary
An Inter-Chassis Link (ICL) failure management system includes a first switch device and a second switch device coupled together by an Inter-Chassis Link (ICL) that is included in a control network and that is configured to provide a primary control channel for transmitting control information between the first switch device and the second switch device. A third switch device is coupled to each of the first switch device and the second switch device by a first aggregated link. When the first switch device determines that the ICL is unavailable, it causes a first port and a second port that provide the first aggregated link to be added to the control network to provide a primary backup control channel. The first switch device may then send control information to the second switch device through the primary backup control channel via the first port and the second port.