Overview

Though not a true protocol, routers and hosts (nodes) use Neighbor Discovery (ND) messages to determine the link-layer addresses of neighbors that reside on attached links and to overwrite invalid cache entries. Hosts also use ND to find neighboring routers that can forward packets on their behalf.

In addition, nodes use ND to actively track the ability to reach neighbors. When a router (or the path to a router) fails, nodes actively search for alternatives to reach the destination.

IPv6 Neighbor Discovery corresponds to a number of the IPv4 protocols — ARP, ICMP Router Discovery, and ICMP Redirect. However, Neighbor Discovery provides many improvements over the IPv4 set of protocols. These improvements address the following:

• Router discovery—How a host locates routers residing on an attached link.
• Prefix discovery—How a host discovers address prefixes for destinations residing on an attached link. Nodes use prefixes to distinguish between destinations that reside on an attached link and those destinations that it can reach only through a router.
• Parameter discovery—How a node learns various parameters (link parameters or Internet parameters) that it places in outgoing packets.
• Address resolution—How a node uses only a destination IPv6 address to determine a link-layer address for destinations on an attached link.
• Next-hop determination—The algorithm that a node uses for mapping an IPv6 destination address into a neighbor IPv6 address (either the next router hop or the destination itself) to which it plans to send traffic for the destination.
• Neighbor unreachability detection—How a node determines that it can no longer reach a neighbor.
• Duplicate address detection—How a node determines whether an address is already in use by another node.