IPv6 Autoconfiguration


Autoconfiguration is a new feature introduced with IPv6 that provides the facility for assigning an IPv6 address to a network interface without manual intervention by the administrator, as is currently necessary with IPv4. There are two types of autoconfiguration, stateless and stateful:

  • Stateless This mechanism enables a host to generate its own IPv6 address based on the ethernet address of the network interface being configured and advertisement information provided by local routers. The routers advertise the format prefixes for subnetworks on the local network and the host itself calculates an interface identifier, based on the ethernet address, to create a unique address for the interface. The IPv6 address is created through a combination of the interface identifier (discussed in the next section) and the prefix information provided by local routers.

    graphics/note_icon.gif

    When autoconfiguring an IPv6 address, if there are no local routers, then the only type of IPv6 address that can be assigned is the link-local address. The interface will be able to communicate only with other hosts on the same network.


  • Stateful This mechanism enables a host to obtain an address and other configuration information from a server, such as a Dynamic Host Configuration Protocol (DHCP) server. The server providing the addresses keeps track of which addresses have been assigned and to whom. It is not the preferred method of autoconfiguration because it requires additional resources to assign an IPv6 address.

graphics/alert_icon.gif

A combination of stateful and stateless autoconfiguration can be used in the setup of an IPv6 interface. The stateless mechanism can be used for a host to configure its own interface, then stateful autoconfiguration can be used to obtain further information, such as format prefixes.


How to Calculate the Interface Identifier

The interface identifier provides the lower 64 bits of an autoconfigured IPv6 address. It is based on the ethernet address of the interface being configured. This section uses the ethernet address 08:00:20:b3:41:53 as the basis for creating the interface identifier.

The ethernet address is composed of two parts , each of 24 bits, where

  • 08:00:20 is the company identifier (CID) assigned to Sun Microsystems for the majority of SPARC systems.

  • b3:41:53 is the vendor-supplied identifier (VID) uniquely assigned by Sun Microsystems.

To calculate the interface identifier, the following steps are performed:

  1. Obtain the ethernet address. This address can be found by running the ifconfig -a command as user root .

  2. Convert the ethernet address to binary. Figure 7.2 shows the converted binary address as well as the hexadecimal representation.

    Figure 7.2. Converted ethernet address.

    graphics/07fig02.gif

  3. Swap bit 7 (from the left). This is the universal/local bit and is changed to create a locally assigned ethernet address. If it's currently a zero, change it to a 1, or vice versa. Figure 7.3 shows the modified address; bit 7 is shown in bold.

    Figure 7.3. Ethernet address with bit 7 swapped.

    graphics/07fig03.gif

  4. Two additional octets need to be inserted into this address between the CID and the VID. These are FF and FE hexadecimal values. Figure 7.4 shows the address in both hexadecimal and binary, with these two octets added.

    Figure 7.4. Ethernet address with two octets added.

    graphics/07fig04.gif

  5. The created interface identifier needs to be converted back to hexadecimal. Also, add the colon (:) characters as separators. This will give a final interface identifier of

     0a00:20ff:feb3:4153 

The unique interface identifier forms only 64 bits of the 128-bit IPv6 address. If this was a link-local IPv6 address, using the fec format prefix, the final autoconfigured IPv6 address would be

 fec0:0000:0000:0000:0a00:20ff:feb3:4153 

Duplicate Address Detection

One of the features of autoconfiguration is that of a duplicate address detection facility. This runs before an IPv6 address is assigned to ensure the address is unique. Duplicate detection checking is carried out regardless of whether stateful or stateless autoconfiguration is used to assign the IPv6 address.

The procedure works by sending a neighbor solicitation message to the network on which the address is going to be assigned. A neighbor advertisement will be received back from any systems that are using the address in question. If no response is received, it is assumed that the address is available to be used, and the assignment takes place.



Solaris 9 Network Administration Exam Cram 2 (Exam Cram CX-310-044)
Solaris 9 Network Administrator Exam Cram 2 (Exam CX-310-044)
ISBN: 0789728702
EAN: 2147483647
Year: 2003
Pages: 174
Authors: John Philcox

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net