IPv6 How to replace ARP with neighbor requests and neighbor announcements.

This article is mainly about the working principle, but also my own doubt, in the IPV4, host sent ARP broadcast to the network with the net segment and all members of this VLAN, I want to find 1.1.1.1, please 1.1.1.1 received a request to reply me.

Then 1.1.1.1 this time to respond to a single broadcast back.

The last host's ARP cache table learns the IP address and Mac on the end, and indicates whether it is dynamic or static learning.

This time the host will be able to communicate with the end-to-end, because I have you in my address book. You have me, too. Learning is what they are to each other.

So how is IPv6 to realize this stuff?

As you can see from the table above, the ICMPV6 type 135 and type=136 are actually used instead of the IPv4 ARP principle.

The following will be combined with the principle of the experiment to carry out in-depth analysis of this process.

Neighbor Request message Ns:icmpv6 type = 135

Neighbor Bulletin Information na+ The combination of solicited node multicast address ff02::1:ffxx:xxxx: ICMPv6 type = 136

How do neighbors ' requests and neighbors ' announcements work? Principle

I personally understand that this place is actually used ICMPV6 type 135 NS, instead of the previous ARP broadcast, I made a comparison table below. About this.

Review IPv6 's "ARP" negotiation principle again: