BFD for router forwarding Fault Detection

In terms of IP network design, faults cannot be restored within less than one second. However, applications such as VoIP are increasingly demanding for rapid Fault Detection and recovery. A new protocol called Bidirectional Forwarding Detection (BFD) will help solve this problem and improve fault detection and recovery speed. As a draft IETF standard, BFD provides a simple method for detecting the forwarding capability of links or systems.

Multi-Layer Detection

BFD is developed step by step from the basic transmission technology, so it can detect faults at all layers of the network. It can be used to detect multiple types of transmission correctness, including Ethernet, Multi-Protocol Label Switching (MPLS) paths, common Routing Encapsulation, and IPSec tunnels.

In essence, BFD is a high-speed, independent Hello protocol (similar to those used in routing protocols, such as Open Shortest Path First Protocol (OSPF ), or an intermediate system-to-intermediate system protocol that can be established with links, interfaces, tunnels, routes, or other network forwarding components ).

BFD can establish a peering relationship with adjacent systems. Then, each system monitors the BFD rate from other systems at a negotiated rate. The monitoring rate can be set incrementally in milliseconds.

When a peer system does not receive a predefined number of packets, It infers that the BFD-protected software or hardware infrastructure is faulty, whether the infrastructure is marking the switching path, other types of tunnels, or the switching Ethernet network.

BFD is deployed on the control plane of the router and other systems. The network faults detected by BFD can be recovered from the forwarding plane (for example, in MPLS fast restart Routing) or from the control plane (for example, when BFD is used to speed up the running of the routing protocol ).

Simple Solution

The simplicity of BFD enables it to be used in some forwarding fault detection solutions. For example, an application of VoIP media gateway connected to an IP core through an switched Ethernet network.

This type of applications poses two challenges: 1. Currently, media gateway does not maintain a peering relationship on the IP layer. Therefore, it lacks a visible means to detect faults between the media gateway and the IP edge router. 2. Ethernet cannot notify hosts or routers that a long-distance part of the switching network has a fault. Rapid fault detection is very important for the high availability of the VOIP network. However, it is not easy to quickly detect link faults when an intermediate router exists between the host and the router.

BFD is simple enough to include in the media gateway platform. In the media gateway example, BFD can be used to maintain the connectivity between the gateway and the edge router.

Faults in the intermediate Ethernet segment will be detected by BFD. BFD switches the gateway and router to the redundant path. Once the fault is detected and confirmed, BFD can trigger the problem solving mechanism in all routing, transmission and tunneling systems.

Because BFD's mission is simple and abstract, it can focus on identifying forwarding faults as soon as possible, so that voice, video, and other demanding services can be successfully forwarded. The BFD protocol will enable service providers to provide VOIP and other real-time services on IP addresses at the reliability and availability levels required by customers.

ArticleInput: CSH responsible editor: CSH