Gateway Ranking for Logical Router Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized networks, there is a need for techniques to minimize the effects of failover between logical router implementations to ensure continuous communication between virtual machines on different subnets, as existing methods do not effectively manage gateway failures and recoveries.
Innovation Solution
A method for managing a cluster of gateways that implements logical ports of logical routers, where an administrator specifies a ranking order for gateways, and a network control system generates an ordered list to designate a master gateway and standby gateways, preserving the current master whenever possible to minimize traffic disruption during failures and recoveries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If the highest-ranked gateway is always designated as master when it becomes operational, then the gateway ranking order is strictly maintained, but traffic disruption occurs during master gateway changes
Solution Approach 1:
The system dynamically adjusts the master gateway selection based on operational status. When the highest-ranked gateway is operational, it becomes master; when it fails, the next highest-ranked gateway becomes master. This dynamic adaptation maintains both ranking integrity and traffic continuity by responding to real-time gateway availability.
Solution Approach 2:
The system pre-establishes a ranked order of gateways before failures occur. This preliminary ranking allows for immediate failover to the next available gateway without complex decision-making during failure events, reducing traffic disruption while maintaining the configured ranking structure.
2Reliability
If the current master gateway is preserved during recovery, then traffic disruption is minimized, but the configured ranking order may not be strictly followed
Solution Approach 1:
The system implements dynamic master gateway selection that balances ranking adherence with traffic continuity. Upon gateway recovery, the system evaluates whether switching to the recovered gateway would cause traffic disruption, and only switches if beneficial, thus dynamically adjusting between strict ranking followance and traffic stability.
Solution Approach 2:
The system changes the selection parameter from strict ranking priority to traffic continuity priority based on operational context. When the current master is operational, it is preserved regardless of ranking; when failed, the next ranked gateway is selected. This parameter change resolves the contradiction between ranking order and traffic continuity.
3Reliability
If gateway failover mechanisms are implemented, then reliability during failures is improved, but system complexity increases
Solution Approach 1:
The system pre-configures gateway rankings and failover priorities before failures occur. This preliminary setup eliminates the need for complex real-time decision algorithms during failover events, reducing system complexity while maintaining reliable failover capability. The ranked list serves as a simple lookup table during failure events.
Solution Approach 2:
The gateway management system automatically detects gateway failures and performs failover to the next ranked gateway without human intervention. This self-service capability simplifies the overall system architecture by eliminating the need for manual gateway management while ensuring reliable failover, thus improving reliability without proportionally increasing complexity.
Data Source
AI summary
Some embodiments provide a method for managing a set of forwarding elements. The method receives configuration information for a set of gateways specifying (i) multiple gateways for implementing logical router ports and (ii) a ranking order of the gateways in the set. The method configures a first gateway in the ranking order as a master gateway for the set of gateways. Upon receiving a first notification that the first gateway is not operational, the method configures a second gateway in the ranking order as the master gateway for the set of gateways. Upon receiving a second, subsequent notification that the first gateway has resumed operation, the method maintains the second gateway in the ranking order as the master gateway for the set of gateways and configures the first gateway in the ranking order as a first standby gateway.