Understanding IPV6 (Part 2)

By: Daniel Imbellino
Updated: Feb 28, 2013

There are also” link local unicast addresses” and unique local unicast addresses”. These address types are used for internal network communications. They are really the equivalent of the “Automatic Private IP address” scheme used in Windows PC’s. APIPA is used to configure workstations on an IPv4 internal network with an address for internal communications with other workstations on the same subnet (router interface). And, just as with IPv4, Link local and unique local addresses cannot be used on the open internet. In IPv4 APIPA addresses consist of an address that begins with 169.254.

Link Local Unicast addresses begin with the fe80: prefix, which consists of the first 10 bits of the address space. The next 64 bits are used for the subnet ID, which are actually not used and left blank with leading zeros, since link local addresses cannot communicate outside their individual subnets. Remember, they are private IP address spaces. The last 54 bits of the address space are used for the interface ID. When we refer to “Link” in link local, or unique local, we are referring to the subnet that a workstation resides on. Workstations with these addresses can communicate with other workstations that share the same link (subnet)only. Link local unicast addresses are now considered deprecated and should be replaced with Unique local unicast addresses.

Unique Local Unicast addresses begin with a site local prefix of FD00: which consists of the first 7 bits of the address space. The 8th bit is set to 1 indicating that this is a unique local address. The next 40 bits are used to represent the Global ID, which is infact a randomly generated value used to identify a specific site within an organization. The next 16 bits identify the subnet ID, which can be used to manage routing of data from one link to another (one subnet to another), something that link local addresses could not do. The last 64 bits identify the workstations network interface card, and this value is almost always derived from the MAC address of that workstations NIC. So with Unique local unicast addresses, you can communicate across multiple internal links, but with Link local unicast addresses you cannot communicate across multiple links, only the workstations on your current link (subnet).

IPv6 address prefix types used to identify different types of addresses:
Global unicast addresses: 2000:
Link-local unicast addresses: fe80:
Unique local unicast addresses: fd00:
Multicast addresses: ff00:

Just as you have a loop back address of in IPv4 that is used to identify the local host (to test your pc’s network interface), IPv6 also has a local host function. This local host function consists of ::1 to represent the workstation at hand.

As we stated before, an IPv6 address consists of 128 bits. Multicast IPv6 addresses are generally used to send data to all workstations located on the same link (subnet). The first 8 bits in a multicast address always start with FF as the address prefix. The following 4 bits identify the “flag bits”, Although 3 of these 4 flag bits are not used and are set to 0. The fourth flag bit is what we call a “transient bit”. This bit specifies whether or not an address is temporary or permanent. If the flag bit is set to 1, then the IPv6 address is temporary. If the flag bit is set to 0, then the IPv6 address is a permanent address. The next following 4 bits in a multicast address are called the “scope bits”. They are used to identify the scope in which the particular multicast address is being used. Because we have 4 bits, we have only 2 to the 4th or 16 possible scope values that can be assigned to a multicast address. Keep in mind, multicast addresses always start with FF as their prefix.

Continue to Understanding IPv6 - Part Three...