While version 6 of the Internet Protocol is similar to version 4, there are significant differences in the operation of the former as compared to the latter. While the primary emphasis of the current ROUTE exam is centered on routing for IPv6 networks, it is still important to have a fundamental understanding of the different protocols and mechanisms used by IPv6. The following IPv6 protocols and mechanisms are described in this section:
- ICMP for IPv6
- The IPv6 Neighbor Discovery Protocol
- IPv6 Stateful Autoconfiguration
- IPv6 Stateless Autoconfiguration
NOTE: As previously stated, the current ROUTE exam is centered on IPv6 routing; therefore, while additional IPv6 protocols and mechanisms are described in this section, configurations pertaining to these protocols and mechanisms will not be included as they are beyond the scope of the current ROUTE exam.
ICMP for IPv6
ICMP is used to reports errors and other information to the source hosts regarding the delivery of IP packets to the intended destination. ICMPv6, which is defined in RFC 2463 as protocol number 58, supports messages for ICMPv4 and includes additional messages for ICMPv6. ICMPv6 is used in the Next Header field of the basic IPv6 packet header. Unlike in IPv4, IPv6 views ICMPv6 as an upper-layer protocol, such as TCP, for example, which means that ICMPv6 is placed after all possible extension headers in the IPv6 packet. The fields that are contained within the ICMPv6 packet are illustrated in Figure 8-12:
Within the ICMPv6 packet header, the 8-bit Type field is used to indicate or identify the type of ICMPv6 message. This field is used to provide both error and informational messages. Table 8-9 lists and describes some common values that can be found within this field:
| ICMPv6 Type | Description |
|---|---|
| 1 | Destination Unreachable |
| 2 | Packet Too Big |
| 3 | Time Exceeded |
| 128 | Echo Request |
| 129 | Echo Reply |
NOTE: It should be noted that these same message types are also used in ICMPv4.
Following the Type field, the 8-bit Code field provides details pertaining to the type of message sent. Table 8-10 illustrates common values for this field, which are also shared by ICMPv4:
| ICMPv6 Type | Description |
|---|---|
| 0 | Echo Reply |
| 3 | Destination Unreachable |
| 8 | Echo |
| 11 | Time Exceeded |
Following the Code field, the 16-bit Checksum field contains a computed value used to detect data corruption in ICMPv6. And finally, the Message or Data field is an optional, variable-length field that contains the data specific to the message type indicated by the Type and Code fields. When used, this field provides information to the destination host. ICMPv6 is a core component of IPv6. Within IPv6, ICMPv6 is used for the following:
- Duplicate Address Detection (DAD)
- The replacement of ARP
- IPv6 Stateless Autoconfiguration
- IPv6 Prefix Renumbering
- Path MTU Discovery (PMTUD)
NOTE: Of the options printed above, DAD and stateless Autoconfiguration will be described later in this section. PMTUD is beyond the scope of the current ROUTE exam and will not be described in any additional detail in this chapter or in the remainder of this guide.
The IPv6 Neighbor Discovery Protocol
The IPv6 Neighbor Discovery Protocol (NDP) is defined in RFC 2461 and is an integral part of IPv6. NDP operates in the Link Layer and is responsible for discovery of other nodes on the link, determining the link layer addresses of other nodes, finding available routers, and maintaining reachability information about the paths to other active neighbor nodes. NDP performs functions for IPv6 similar to the way ARP and ICMP Router Discovery and Router Redirect protocols do for IPv4. However, it is important to remember that NDP provides greater functionality than the mechanisms used in IPv4. Used in conjunction with ICMPv6, NDP allows for the following:
- Dynamic Neighbor and Router Discovery
- The replacement of ARP
- IPv6 Stateless Autoconfiguration
- Router Redirection
- Host Parameter Discovery
- IPv6 Address Resolution
- Next-Hop Router Determination
- Neighbor Unreachability Detection (NUD)
- Duplicate Address Detection
NOTE: You are not required to delve into specifics on each of the advantages listed above.
NDP defines five types of ICMPv6 packets, which are listed and described in Table 8-11:
| ICMPv6 Type | Message Type Description and IPv6 Usage |
|---|---|
| 133 | Used for Router Solicitation (RS) messages |
| 134 | Used for Router Advertisement (RA) messages |
| 135 | Used for Neighbor Solicitation (NS) messages |
| 136 | Used for Neighbor Advertisement (NA) messages |
| 137 | Router Redirect |
RS messages are sent by hosts when interfaces are enabled for IPv6. These messages are used to request that routers on the local segment generate RA messages immediately, rather than at the next scheduled RA interval. Figure 8-13 shows a wire capture of an RS message:
Upon receiving the RS message, advertise their presence using RA messages which typically include prefix information for the local link as well as any additional configuration, such as suggested hop limits. The information contained with the RA is shown in Figure 8-14:
IPv6 NS messages are Multicast by IPv6 routers on the local network segment and are used to determine the Data Link address of a neighbor or verify that a neighbor is still reachable. These messages are also used for Duplicate Address Detection. While delving into detail on NS messages is beyond the scope of the ROUTE exam, Figure 8-15 shows a wire capture of an IPv6 Neighbor Solicitation message:
NA messages are typically sent by routers on the local network segment in response to received NS messages. However, if, for example, an IPv6 prefix changes, routers may also send out unsolicited NS messages advising other devices on the local network segment of the change. As is the case with NA messages, while going into detail on the format or fields contained within the NA message is beyond the scope of the ROUTE exam, Figure 8-16 shows a wire capture of the Neighbor Advertisement message which is also sent via IPv6 Multicast:
And finally, router redirection uses ICMPv6 Redirect messages, which are defined as message type 137. Router redirection is used to inform network hosts that a router with a better path to the intended destination exists on the network. It works in the same manner as ICMPv4 redirects used in current IPv4 networks.
IPv6 Stateful Autoconfiguration
As previously stated earlier in this chapter, stateful Autoconfiguration allows network hosts to receive their addressing information from a network server, e.g. via DHCP. This method of Autoconfiguration is supported by both IPv4 and IPv6. In IPv6 networks, DHCPv6 is used to provide stateful Autoconfiguration services for IPv6 hosts. In IPv6 implementations, when an IPv6 host receives RA messages from routers on the local network segment, the host examines these packets to determine if DHCPv6 can be used. The RA messages provide this information by setting either the M or the O bits on.
The M bit in Router Advertisement messages is the Managed Address Configuration Flag bit. When this bit is set, i.e. contains a value of 1, it instructs the IPv6 host to obtain a stateful address, which is provided by DHCPv6 servers. The O bit in Router Advertisement messages is the Other Stateful Configuration Flag bit. When this bit is set, i.e. contains a value of 1, it instructs the IPv6 host to use DHCPv6 to obtain more configuration settings, such as DNS and WINS servers, for example. While one of the advantages of IPv6 is stateless Autoconfiguration capability, stateful Autoconfiguration still provides several advantages which include:
- Greater controls than those provided by Stateless Autoconfiguration
- It can be used on networks when Stateless Autoconfiguration is available
- It provides addressing to network hosts in the absence of routers
- It can be used for network renumbering by assigning new prefixes to hosts
- It can be used to issue entire subnets to customer premise equipment
IPv6 Stateless Autoconfiguration
In IPv6, stateless Autoconfiguration allows hosts to configure their Unicast IPv6 addresses by themselves based on prefix advertisements from routers on the local network segment. There are three main mechanisms that allow for stateless Autoconfiguration in IPv6:
- Prefix Advertisement
- Duplicate Address Detection
- Prefix Renumbering
IPv6 prefix advertisement uses ICMPv6 Router Advertisement messages, which are sent to the all-hosts-on-the-local-link IPv6 Multicast address FF02::1. By design, only routers are allowed to advertise prefixes on the local link. When Stateless Autoconfiguration is employed, it is imperative to remember that the prefix length used must be 64-bit, for example: 2001:1a2b::/64.
Following the configuration of the prefix, Router Advertisement messages used for IPv6 stateless Autoconfiguration include the following information:
- The IPv6 prefix
- The lifetime
- The default router information
- Flags and/or Options
As previously stated, the IPv6 prefix must be 64-bit. In addition to this, multiple IPv6 prefixes may be advertised on the local segment. When hosts on the network segment receive the IPv6 prefix, they append their MAC address to the prefix in EUI-64 format, which was described earlier in this chapter, and automatically configure their IPv6 Unicast address. This provides a unique 128-bit IPv6 address to each host on the network segment.
The lifetime value for each advertised prefix is also provided to the nodes and may contain a value from 0 to infinite. When nodes receive the prefix, they validate the lifetime value and cease using the prefix when the lifetime value reaches 0. Alternatively, if a value of infinite is received for a particular prefix, the network hosts will never cease using that prefix. Each advertised prefix contains two lifetime values: the valid lifetime and the preferred lifetime.
The valid lifetime is used to determine how long the host address will remain valid. When this value expires, i.e. reaches a value of 0, the host address becomes invalid. The preferred lifetime is used to determine the duration for how long an address configured via Stateless Autoconfiguration will remain valid. This value must be less than or equal to the value specified in the valid lifetime, and is typically used for prefix renumbering.
The default router provides information about the existence and lifetime of the default router’s IPv6 address. By default, the default IPv6 address used for default routers is the Link-Local address (FE80::/10). This allows the global Unicast address to be changed without interrupting network services, as would be the case in IPv4 if a network was renumbered. And finally, the Flags and Options fields can be used to instruct network hosts to use Stateless Autoconfiguration or Stateful Autoconfiguration. These fields are included in the wire capture of the Router Advertisement shown in Figure 8-14.
Duplicate Address Detection (DAD) is an NDP mechanism used in stateless Autoconfiguration when a host on the network segment is booting up. DAD mandates that before a network host permanently configures its own IPv6 during boot up, it should validate that the IPv6 address that it wants to use is not already being used by another network host.
DAD performs this by using Neighbor Solicitation (ICMPv6 Type 135) and Solicited-Node Multicast addresses. The host sends a Neighbor Solicitation on the local network segment using an unspecified IPv6 address, i.e. the :: address, as its source address and the Solicited-Node Multicast address of the IPv6 Unicast address it wants to use as the destination address. If no other host is using this same address, the host will not automatically configure itself with this address; however, if no other device is using the same address, the host automatically configures itself and begins to use this IPv6 address.
And finally, prefix renumbering allows for the transparent renumbering of network prefixes in IPv6 when changing from one prefix to another. Unlike in IPv4 where the same global IP address can be advertised by multiple providers, the strict aggregation of the IPv6 address space prevents providers from advertising prefixes that do not belong to their organization.
In cases where a transition is made from one IPv6 Internet provider to another, the IPv6 prefix renumbering mechanism provides a smooth and transparent transition from one prefix to another. Prefix Renumbering uses the same ICMPv6 messages and Multicast address used in prefix advertisement. Prefix renumbering is made possible by using the time parameters contained within the Router Advertisement messages.
In Cisco IOS software, routers can be configured to advertise current prefixes with the valid and preferred lifetime values decreased to a value closer to zero, which allows those prefixes to become invalid faster. The routers are then configured to advertise the new prefixes on the local network segments. This allows the old and new prefixes to exist on the same network segment.
During this transition period, hosts on the local network segment use two Unicast addresses: one from the old prefix and one from the new prefix. Any current connections using the old prefix are still handled; however, any new connections from these hosts are made using the new prefix. When the old prefix expires, only the new prefix is used.