Backup Server Failover via Discovery Message Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In failover systems, network failures can lead to impaired services for clients when communication breaks between a relay agent and the main server, as the backup server remains unresponsive until it detects the main server's unresponsiveness, causing service disruptions.
Innovation Solution
Implementing a method where the backup server becomes responsive to discovery messages if it receives a predetermined number of discovery messages from a relay agent and if a time threshold is exceeded since the last valid discovery message, allowing it to take over service requests and ensure continuous client service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the backup server remains unresponsive to service requests in normal operation, then service request handling is simplified and main server control is maintained, but service continuity is impaired when network failure occurs between the relay agent and main server
Solution Approach 1:
The backup server performs preliminary monitoring of discovery messages from relay agents and prepares to take over before actual failure occurs. By tracking message counts and time thresholds in advance, the backup server can immediately respond when the main server becomes unreachable, ensuring service continuity without complex real-time decision-making during failure events.
Solution Approach 2:
The system implements feedback through the backup server continuously monitoring discovery messages from relay agents. When the main server fails to respond, the backup server detects this through the absence of expected responses and automatically activates, creating a closed-loop feedback mechanism that maintains service continuity based on real-time system state.
2Reliability
If the backup server immediately takes over when it detects main server unresponsiveness, then service continuity is improved, but false activations may occur due to temporary network glitches
Solution Approach 1:
The backup server sets up preliminary thresholds for activation: a minimum number of discovery messages (N) must be received without response from the main server, and a time threshold (T) must be exceeded. These pre-configured parameters prevent immediate reaction to transient network issues while ensuring timely failover for genuine failures.
Solution Approach 2:
The system uses parameter changes (message count N and time threshold T) to transform the backup server's state from passive to active. By monitoring these parameters over time, the system distinguishes between temporary network glitches and genuine failures, activating failover only when parameters indicate sustained unresponsiveness.
3Measurement precision
If the backup server monitors every discovery message from relay agents, then detection accuracy is improved, but communication overhead and processing load increase
Solution Approach 1:
The backup server performs partial monitoring by counting discovery messages rather than fully processing each message content. It only activates when the count reaches threshold N within time T, performing excessive monitoring during normal operation but minimal processing during failure detection, optimizing the balance between accuracy and energy consumption.
Data Source
AI summary
A method for providing services by a pair of servers to clients through one or more relay agents. The pair includes a second server configured to respond to discovery messages, and a first server configured to be unresponsive to the discovery messages. The method includes: at the first server, receiving a first discovery message from a first relay agent at a first time; and configuring the first server to become responsive to the first discovery message when a number of discovery messages from the first relay agent is equal to or greater than a predetermined number threshold and when a time lapse between the first time and a second time is equal to or greater than a predetermined time threshold. The count and time period are reset if the second server is determined to be responsive to discovery messages by monitoring the client request messages.