EVPN Multihoming Packet Forwarding After DF Link Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In EVPN multihoming networking, when a CE device does not support the connectivity fault management (CFM) protocol, link faults between the designated forwarder (DF) device and the CE device go undetected, leading to interruptions in uplink traffic.
Innovation Solution
A method where the first PE device sends a packet to a second PE device to indicate it to forward the packet to the CE device, ensuring the CE device updates its MAC entry, thereby maintaining normal uplink traffic flow by switching the DF device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the CE device does not support the CFM protocol, then the device complexity is reduced, but the reliability of link fault detection deteriorates
Solution Approach 1:
The patent introduces an intermediary mechanism where the PE device acts as a mediator to detect link faults. Instead of requiring the CE device to detect faults directly (which would increase its complexity), the PE device monitors the link status and triggers DF device switching when faults are detected, thereby maintaining high reliability without increasing CE device complexity
Solution Approach 2:
The patent implements preliminary action by pre-configuring multiple DF devices and establishing backup relationships before faults occur. When a link fault is detected, the pre-established backup DF device can immediately take over, ensuring rapid fault response without requiring complex real-time decision-making at the CE device
2Device complexity
If the DF device switching is delayed, then the device complexity is reduced, but the productivity of uplink traffic forwarding deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the PE device continuously monitors link status and provides real-time feedback about fault conditions to the control system. This automatic feedback loop enables rapid DF device switching without complex manual intervention, maintaining high uplink traffic forwarding efficiency while keeping device control complexity manageable
Solution Approach 2:
The system performs self-service through automated DF device switching based on link fault detection. The control system automatically triggers the switching process when faults are detected, eliminating the need for complex manual intervention and enabling rapid response that maintains high forwarding productivity
3Device complexity
If the CE device continues to send packets to the original DF device, then the device complexity is reduced, but the reliability of uplink traffic transmission deteriorates
Solution Approach 1:
The PE device serves as an intermediary that monitors link faults and manages DF device switching. When a fault is detected, the PE device triggers the switching process, and the CE device simply follows the updated forwarding information without needing to make complex decisions about whether to switch, thereby maintaining transmission reliability without increasing CE device complexity
Data Source
AI summary
This application provides a packet sending method, a device, and a system, and belongs to the field of EVPN technologies. In the solutions provided in this application, when a DF device switches from a first PE device to a second PE device due to a fault in a link between the first PE device and a CE device, the first PE device may send a first packet to the second PE device, so that the second PE device forwards the first packet to the CE device. Therefore, the CE device may generate or update a MAC entry in time based on the first packet forwarded by the second PE device, and may send a packet to the second PE device based on the MAC entry. In this way, the second PE device that switches to the DF device may forward the packet from the CE device, so that normal forwarding of uplink traffic sent from the CE device can be ensured, and interruption of the uplink traffic of the CE device can be avoided.