ICR Node State Transition for Standby Failure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In network systems using Border Gateway Protocol (BGP), active Inter-Chassis Redundancy (ICR) nodes fail to detect standby node failures, leading to network service failures and degradation due to suppressed communications about non-best paths, resulting in unacceptable operations and revenue loss.
Innovation Solution
Implementing a network node with a microprocessor that transitions from an active-with-peer to an active-without-peer state upon detecting a standby ICR node's partial or complete non-functionality, preventing switchover and ensuring continued network traffic processing and forwarding by establishing a dedicated ICR transport channel for heartbeat and status messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If BGP routing protocol is used for inter-autonomous system routing, then network reachability information can be exchanged between BGP hosts, but the active ICR node cannot detect standby node failures due to suppressed communications about non-best paths
Solution Approach 1:
The patent introduces a dedicated ICR transport channel as an intermediary communication path between active and standby ICR nodes. This separate channel bypasses the BGP routing protocol's path suppression mechanism, allowing failure detection messages to be transmitted independently of best-path routing decisions. The intermediary channel ensures that reliability information can be exchanged without being blocked by the routing protocol's normal operation.
2Reliability
If ICR switchover is implemented for high availability, then network service can be maintained during active node failure, but network service failures occur when switchover is attempted to a non-functional standby node
Solution Approach 1:
The patent implements a feedback mechanism where the active ICR node continuously monitors the standby node's functionality through heartbeat messages and status indications transmitted over the dedicated ICR transport channel. Before executing a switchover operation, the active node receives feedback confirming the standby node's operational status. This feedback loop prevents switchover to non-functional standby nodes, eliminating the harmful effect of service failures from invalid switchover attempts.
Solution Approach 2:
The patent performs preliminary failure detection and status verification of the standby node before initiating switchover operations. By using the dedicated ICR transport channel to exchange status messages in advance, the system determines whether the standby node is functional before committing to a switchover. This preliminary action prevents premature or invalid switchover decisions that would otherwise cause service failures.
3Productivity
If BGP path suppression is used to maintain best-path routing, then routing efficiency is improved, but the active ICR node remains unaware of standby node presence and status
Solution Approach 1:
The patent segments communication functions into two separate channels: the BGP routing protocol channel for best-path routing information exchange, and a dedicated ICR transport channel for standby node status and failure detection information. This segmentation allows BGP path suppression to continue optimizing routing efficiency while the separate ICR channel independently carries status information without being affected by path suppression mechanisms.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A network node for routing messages in a communications system, includes at least one network interface that communicates messages with a plurality of other network nodes, and a microprocessor that sends and receives messages. The microprocessor transitions from operation in an active-with-peer state to operation in an active-without-peer state in response to detecting that a standby inter-chassis redundancy (ICR) node has become or will become at least partially non-functional as a backup message router for the network node. While operating in the active-with-peer state, the microprocessor responds to a switchover triggering event by switching-over at least part of its message routing responsibility to the standby ICR node. In contrast, while operating in an active-without-peer state, the microprocessor does not switchover routing responsibility to the standby ICR node in response to the switchover triggered event.