Bidirectional Forwarding Detection (BFD) for Multi-point Networks and Virtual Router Redundancy Protocol (VRRP) Use Case

The is the current specification of the Virtual Router Redundancy Protocol (VRRP) for IPv4 and IPv6 networks. VRRPv3 allows for faster switchover to a Backup router. Using such capability with software-based implementation of VRRP is may prove challenging. But it still may be possible to deploy VRRP and provide sub-second detection of Master router failure by Backup routers. Bidirectional Forwarding Detection (BFD) had been originally defined detect failure of point-to-point (p2p) paths: single-hop , multihop . Single-hop BFD may be used to enable Backup routers to detect failure of the Master router within 100 msec or faster. demonstrates how, with some extensions to , that can be achieved. extends for multipoint and multicast networks, which is precisely characterizes deployment scenarios for VRRP over LAN segment. This document demonstrates how point-to-multipoint (p2mp) BFD can enable faster detection of Master failure and thus minimize service disruption in a VRRP domain. The document also defines the extension to VRRP to bootstrap a VRRP Backup router to join in p2mp BFD session.

BFD: Bidirectional Forwarding Detection p2mp: Pont-to-Multipoint VRRP: Virtual Router Redundancy Protocol

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here.

A router may be part of several Virtual Router Redundancy groups, as Master in some and as Backup in others. Supporting sub-second mode for VRRPv3 for all these roles without specialized support in data plane may prove to be very challenging. BFD already has many implementations based on HW that are capable to support multiple sub-second session concurrently.

may provide the efficient and scaleable solution for fast-converging environment that uses default route rather than dynamic routing. Each redundancy group presents itself as p2mp BFD session with its Master being the root and Backup routers being tails of the p2mp BFD session. displays the extension of VRRP to bootstrap tail of the p2mp BFD session. Master

where new fields are interpreted as: B(FD) - one bit flag that indicates that the Master Discriminator field is appended to VRRP packet defined in ; Master Discriminator - My Discriminator value allocated by the root of the p2mp BFD session. The Master router that is configured to use p2mp BFD to support faster convergence of VRRP starts transmitting BFD control packets with VRID as source IP address and My Discriminator. The same value of My Discriminator MUST be set as value of Master Discriminator field and BFD flag MUST be set in the VRRP packet. Backup router demultiplexes p2mp BFD test sessions based on VRID that it been configured with and the My Discriminator value it learns from the received VRRP packet. When a Backup router detects failure of the Master router it re-evaluates its role in the VRID. As result, the Backup router may become the Master router of the given VRID or continue as a Backup router. If the former is the case, then the new Master router MUST select My Discriminator and start transmitting p2mp BFD control packets using Master IP address as source IP address for p2mp BFD control packets. If the latter is the case, then the Backup router MUST wait for VRRP packet from the new VRRP Master router that will bootstrap new p2mp BFD session.

This document makes no requests for IANA allocations. This section may be deleted by RFC Editor.

Security considerations discussed in , , and , apply to this document.