Understanding some tricky IPv6 Addresses

The first good thing about IPv6 is that you have an incredible number of addresses for each individual on planet Earth ( assuming a population of 10 billion, there would be something around 34,000 trillions of trillions of addresses per person !!) On the other hand, for someone that has worked with IPv4 all life long, understanding some of the addresses involved in basic v6 operations may be challenging.

Figure 1 brings, as a reference, the summarized output of the classic show ip interface command. The example highlights that v4 uses broadcast addresses, a concept that was not integrated into v6 definitions. To compensate for the absence of broadcast addresses, v6 employs various types of multicast group addresses to perform tasks such as neighbor discovery and duplicate address detection.

Figure 1: Summary output of the “show ip interface” command

Figure 2 documents a typical instance of the show ipv6 interface command, which reveals some interesting facts:

  • It underlines that IPv6 does not employ broadcast addresses.
  • It displays the link-local address, which are used in activities like automatic address configuration and neighbor discovery, or in situations when no routers are present. These addresses are, most of the time, automatically generated by combining the link-local prefix, FE80::/10, with a sequence of 54 zeros and a 64-bit interface identifier. As the name implies this class of addresses is not routed. They are significant only on the local link.
  • It shows the global unicast address, which is a routable IPv6 address. This address may be defined statically or be derived from some sort of auto-configuration process.
  • It reveals that the router has joined two reserved multicast group addresses of link-local scope, FF02::1 (all-nodes) and FF02::2 (all routers).
  • It documents the existence  of two solicited-node multicast addresses, which appear in messages used for tasks such as layer 2 address resolution. The solicited-node multicast address of a node has a link-local scope and is derived by juxtaposing the lower 24 bits of its unicast address to the prefix FF02::1:FF/104.

A question may arise at this point: what are these two solicited-node multicast addresses ?

  • One is associated with the link-local unicast address (dynamically generated from the MAC)
  • The second relates to the global unicast address (statically configured as 2001:db8:49::2/64)

Figure 2: Sample output of the “show ipv6 interface” command

 ** Topics for study:

  • Revisit the previous post to learn more about the EUI-64 interface ID.
  • Research and list some other reserved multicast addresses. What they are used for ?
  • Can you describe the concept of address scope in IPv6 ?
  • Consider that a router has the MAC address 001f.9e56.e4c8 on its interface FastEthernet3 and that it has not been assigned any unicast address. What would be its link-local address ? What reserved multicast-groups will the router join (on this interface) ? How many solicited-node multicast group addresses will the router use ? (What are these addresses ?)

** Related Posts:


Leave a comment

Filed under English, IPv6

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s