BGP IPv6 Capability Pre-Negotiation for Stable Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of enabling IPv6 capabilities in Border Gateway Protocol (BGP) networks results in extensive traffic loss due to re-establishment processes, particularly in large networks with numerous nodes, as existing methods require renegotiation of capabilities, leading to downtime and disruption.
Innovation Solution
A method for deploying IPv6 routing that involves advertising a new address-family capability alongside existing capabilities, filtering out IPv6 routes initially, and then undoing BGP filters to allow route exchange when IPv6 is enabled, thereby avoiding the need for re-establishment processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If BGP IPv6 capabilities are negotiated separately from IPv4 capabilities requiring re-establishment process, then IPv6 capability is properly enabled, but extensive traffic loss occurs during the transition
Solution Approach 1:
The patent applies preliminary action by pre-negotiating BGP IPv6 capabilities during the initial BGP session establishment or during a maintenance window when traffic impact is minimal. The system prepares IPv6 route filtering rules and capability advertisements in advance, so that when IPv6 is actually enabled, the BGP peering relationship is already established and ready to carry IPv6 traffic without requiring session re-establishment that would cause traffic loss.
2Adaptability or versatility
If BGP peers go through re-establishment process when enabling IPv6, then new IPv6 routes can be installed, but BGP routes are deleted from data plan causing traffic to be dropped
Solution Approach 1:
The patent applies segmentation by separating the BGP capability negotiation from the route installation process. It segments the IPv6 enablement into distinct phases: capability advertisement phase, route filtering phase, and route installation phase. This allows the BGP peering relationship to remain stable while IPv6 routes are progressively introduced through controlled filter removal, preventing complete route deletion and maintaining traffic continuity.
Solution Approach 2:
The patent uses route filtering rules as an intermediary mechanism to control the introduction of IPv6 routes. Instead of directly reinstalling routes that would cause instability, the system uses filtering rules to selectively permit or deny IPv6 route installation. This intermediary control allows smooth transition from IPv4-only to dual-stack operation without abrupt route changes that would disrupt traffic.
3Extent of automation
If commit push is sent to all remote firewalls to enable IPv6, then IPv6 is enabled across the network, but BGP sessions are torn down requiring renegotiation
Solution Approach 1:
The patent applies preliminary action by including IPv6 capability advertisements in the initial BGP OPEN message or in capability update messages during a maintenance window, rather than waiting until IPv6 is fully enabled. This preliminary capability negotiation ensures that when the commit push is sent to enable IPv6 on all firewalls, the BGP sessions are already configured to support IPv6 and do not need to be torn down for renegotiation, maintaining session stability during automated network-wide deployment.
Data Source
AI summary
Techniques for deploying IPv6 routing are disclosed. A system, process, and/or computer program product for deploying IPv6 routing includes advertising in Border Gateway Protocol (BGP) a new address-family capability in combination with an existing address-family in a network that supports a plurality of address families, and undoing BGP filters to allow BGP routes to be exchanged at a time that a network administrator enables the new address-family capability in the network.