Lesson 5E

Neighbor Discovery process page 154

1. A host sends a neighbor solicitation message (which is an ICMPv6 packet that has the Type field set to 135) across the local link. This message contains the sender's IPv6 address as the Source Address and the solicited-host multicast address corresponding to the target host's IPv6 address. The neighbor solicitation message also includes the link-layer address of the source host. 2. After receiving the neighbor solicitation message, the target host replies with a neighbor advertisement message (which has the Type field set to 136) on the local link. In this message, the source address is the target's IPv6 address (actually, the IPv6 address of the host interface), and the destination address is the IPv6 address of the host that sent the neighbor solicitation message. The data portion of the neighbor advertisement message includes the link-layer address of the target host. 3. After the source host receives the neighbor advertisement, communication can occur between the source and destination hosts.

IPv6

In CIDR notation addresses can contain up to 128 bits instead of the 32-bit limit in IPv4.

FE80::/10

is reserved for link-local addresses

ff00::/8

is reserved for multicast addresses

IPv6 subnet prefix

is the combination of the site prefix and the subnet ID.

IPv6

uses three different transmission types: unicast, multicast, and anycast

IPv6

• A new field in the IP header of packets enables IP to guarantee the allocation of network resources when requested by time-dependent services such as voice and video transmission.

Neighbor Discovery

• Determine the link-layer address of a host on the local link

Feature of protocol binding

• Dual-stack is the side-by-side implementation of IPv4 and IPv6.

Feature of protocol binding

• In Windows, you can specify the binding order in which to bind protocols to a network interface. • When you set the binding order to prefer the protocol you most frequently use on your network, your system does not attempt to use other protocols to access the network, thus increasing the efficiency of the connection.

IPv6

• Stateless auto-reconfiguration of hosts allows the host to configure automatically when connected to a routed IPv6 network.

IPv4 Address Space Limitations

• The 32-bit IP address space itself, which provides only a theoretical maximum of 2to32 or approximately 4,295 billion, separate addresses.

IPv4 Address Space Limitations

• The depletion of Class A and Class B IP address assignments.

IPv4 Address Space Limitations

• The division of the address space into fixed classes, with the result that host addresses falling either between classes or between subnets are unavailable for assignment.

IPv6 Address

128-bit binary number assigned to a computer on a TCP/IP network. A 128- bit address provides 2128 potential address combinations, which equals 340 trillion trillion trillion (340 undecillion) possible addresses. Not all of those addresses will be publicly routable. Some of the bits in the address represent the network segment; the other bits represent a specific interface on a host. IPv6 addresses are not case-sensitive.

Zone ID

A % sign and a number at the end of an IPv6 address that is used to identify the link the computer belongs to. the number displayed after a percent sign is the zone ID. It is used to identify which network adapter to use for communications. You can also use it to identify network segments. For example, in the address fe80:42b8:9aff:fedc:ce51%18, the zone ID is 18.

IPv6

A contiguous sequence of 16-bit blocks set to 0 in the colon-hexadecimal format can be compressed to :: (known as double-colon). To avoid ambiguity, you can perform the double-colon substitution only once per address.

Neighbor Discovery

A process whereby routers learn about all of the devices on their networks. On IPv4 networks, this process is managed by ARP with help from ICMP. On IPv6 networks, NDP (Neighbor Discovery Protocol) automatically detects neighboring devices and automatically adjusts when nodes fail or are removed from the network.

"Unicast" Link-local

Addresses that are used to communicate and automatically assigned on private network segments with no router. Link-local addresses usually begin with FE80, but can range from FE80 to FCFF. They are the equivalent of self-assigned IPv4 automatic private IP addressing (APIPA) addresses. These are network addresses that are valid only for communications within the network segment or the broadcast domain that the host is connected to. Link-local addresses are usually not guaranteed to be unique beyond a single network segment, so routers do not forward packets with link-local addresses.

Unicast site-local

Addresses used for internal networks that are not routable on the Internet. The equivalent of the IPv4 private, nonroutable address blocks. Site-local addresses begin with FEC0::/10. In Sept 2004, the Internet Assigned Numbers Authority (IANA) requested the IETF deprecate this address type because it was ambiguous and could lead to misconfigurations resulting in network security leaks. Site-local addresses have been replaced with "unique local addresses" in the block fc00::/7. Of that block, fd00::/8 has been set aside for organizations to use in their internal networks.

Unicast IPv6

Addresses used on mixed networks to support routing of IPv6 data across IPv4 networks. This class will be phased out when all routers convert to IPv6. An example of transitional addresses is Microsoft's IPv4- compatible version that Windows sometimes uses during 6-4 transition. In this version, the host ID is not the EUI, but instead the 4 octets of the IPv4 address. So, for example, if the IPv4 address is 192.168.1.26, the IPv6 address is 2001:0db8:85A3:0:192:168:1:26/64. 2002::/16 is a 6-to-4 routing prefix, meaning that networks that are in transition using 6-to-4 will use this prefix.

protocol binding

Assigning a protocol to a network interface card (NIC) As protocols govern data transmission, it is critical to bind the protocol to the network interface as it creates a path for the flow of data. Multiple protocols can be bound to a NIC, and the NIC can use any of the protocols that are bound to it to communicate with other hosts on the network.

IPv6

Even though addresses can be up to 128 bits in length, subnets on MAC layer networks always use 64-bit interface identifiers.

EUI

Extended Unique Identifier aka MAC address A host computer running IPv6 can use an IEEE standard called the extended unique identifier (EUI) to self-assign its interface addresses. With EUI, the MAC address is padded in the center with FFFE, extending it to 64 bits in length. For instance, a MAC address of 01-00-11-22-33-44 would be changed to 0100:11FF:FE22:3344 to become the last 64 bits of the host's IPv6 address. While not a requirement, most vendors are adopting this format, as it's the easiest to implement.

IPv6

For readability, the address is usually separated by colons into eight groups of four hexadecimal digits. Each quartet of four hexadecimal digits equals 16 bits.

Unicast global address

Globally routable public addresses. Also known as aggregatable global unicast addresses, they are designed such that they can be summarized for efficient routing. Global addresses are the equivalent of the entire IPv4 public address space.

IPv6

Goal is to keep the IP headers as small as possible, to make access to the address more efficient and quicker. Non-essential information in IPv6 headers is moved to optional extension headers.

IPv4 vs. IPv6 addresses

IPv4 addresses use 32 bits, whereas IPv6 addresses use 128 bits.

IPv4 vs IPv6 Headers

IPv6 headers are simpler, but expandable if you need to include options.

IPv6 tunneling

IPv6 packets can be encapsulated inside IPv4 datagrams IPv4 and IPv6 are not compatible. One of the ways that IPv6 can be implemented on IPv4 networks is through the tunneling proces

Router Solicitation and Advertising

IPv6 routers regularly advertise information on the links to which they are connected. These advertisements are Internet Control Message Protocol Version 6 (ICMPv6) router advertisement (RA) messages, with the Type field set to 134 and sent to the all-nodes multicast address ff02::1. All the hosts on a link that belong to this group, and hosts configured for autoconfiguration, analyze those messages. Upon reception of an RA message, an autoconfiguring host not already configured with the corresponding global address will prepend the advertised prefix to the unique identifier. Hosts can send router solicitation (RS) messages (ICMPv6 messages with the Type field set to 133) to all routers on the link (ff02::2). Hosts that have not configured an address yet use the unspecified address "::" for the source address of the RS message. The routers respond with an RA message containing the unicast address for the RS sending host.

Feature of protocol binding

In Linux, the ifconfig command will display whether or not you are using IPv6. You can disable IPv6 by editing /etc/sysctl.conf to add the following lines to the file and restart the sysctl service: net.ipv6.conf.all.disable_ipv6 = 1 net.ipv6.conf.default.disable_ipv6 = 1 net.ipv6.conf.lo.disable_ipv6 = 1

2002::/16

Is reserved for 6to4 routing tunneling

2001:db8::/32

Is reserved for use in documentation

IPv6 Interface ID

Last 64 bits (or the last half) assigned to interfaces, not to the host. Must be unique on the same subnet as other IPv6 addresses

IPv4 vs IPv6 loopback address

On an IPv4 network, it is 127.0.0.1, and on an IPv6 host, it is ::1.

RA

Router Advertisement

IPv4 vs IPv6 Default route address (anything unknown is sent to the default gateway)

The IPv4 default route address is 0.0.0.0/0, and the IPv6 default route address is ::/0.

IPv6 prefix

The first numbers in an IPv6 address that identify the type of IPv6 address. For example, it can be a global unicast address (prefix of 2), a link-local address (prefix of fe80), a unique local address (prefix of fd), or a Teredo address (prefix of 2001:000). the leftmost bits of the address that are used for routing IPv6 packets are referred to as the prefix. That is the terminology in IPv6, rather than "network address." The length of the prefix is indicated by using CIDR notation, so in this example address: 2001:0db8:85a3:0000:0000:8a2e: 0370:7334/64

ICMPv6

The version of ICMP used with IPv6 networks. ICMPv6 performs the functions that ICMP, IGMP, and ARP perform in IPv4. It detects and reports data transmission errors, discovers other nodes on a network, and manages multicasting.

IPv6

While all eight groups must have four digits, you can omit any leading zeros from the address notation.

IPv4

addressing scheme that increases the available pool of IP addresses by implementing a 128-bit binary address space. This equates to 340 trillion trillion trillion addresses. IPv6 also includes new features, such as simplified address headers, hierarchical addressing, support for time-sensitive network traffic, required security, and a new structure for unicast addressing.

IPv6 Subnet ID

fourth block of the address • The next 16 bits (0000) after the 48 bits denote the subnet ID

IPv4 Address Space Limitations

• The fact that IP address classes provide a small number of host addresses, leading to difficulty matching IP address leases to a company's needs and IP addresses being wasted.

IPv6 prefix

• The first 48 bits of the IPv6 address (2001:0db8:85a3) denote the network prefix, or site prefix.

Dual-stack

• This is where both protocols run on the same network infrastructure, and there's no need to tunnel IPv6 inside IPv4 or vice versa. • Windows has provided dual-stack support since Windows Vista .

Neighbor Discovery

• Track neighboring devices.

IPv4 Address Space Limitations

• Unassigned and unused address ranges within existing Class A and Class B blocks.

Neighbor Discovery

• Verify that a neighbor host can be reached.

Feature of protocol binding

• When a network interface is bound with multiple protocols, it attempts to connect to a receiving host by sequentially testing the available protocols until it gets a response from the receiving host using a protocol. • This carries an inherent risk that the protocol that the host responds to might not be the most efficient one—it is simply the first compatible protocol in the sender's protocol list that the two hosts have in common.

IPv6

• implements network-layer encryption and authentication with IPSec.