IPv6 address types pART2 IPv6 address types pART2

Editor's Notes

• #1: This is the second of 2 PowerPoint presentations that introduce IPv6. The two presentations combined provide information on the purpose and benefits of IPv6 as well as an explanation of the 128-bit addressing structure. NOTE: Most of the slides have notes that provide additional detail and background information. These can be used for student handouts, if desired, and can also be used as lecture points for the instructor.
• #2: This objectives of this lesson are to provide additional details on IPv6 address types with a focus on unicast and multicast addresses. This lesson describes methods of address assignment including stateless autoconfiguration. IPv6 subnetting and address aggregation are also discussed. The final section provides information on configuration and verification of IPv6 on networking devices.
• #3: This section provides additional details on IPv6 address types with a focus on unicast and multicast addresses.
• #4: As we have seen in Lesson 1, the IPv6 address space is very large. The first four Hexadecimal digits of the prefix help to classify the various blocks of IPv6 addresses. The ability to identify an address type by its prefix is essential to understanding IPv6. In the table, the prefixes highlighted in red are important ones to remember. The largest block of assignable addresses, from Hex 2000 to Hex 3FFF, is known as the aggregatable global unicast range which represents approximately 12.5% of the total IPv6 address space. From this range, only addresses from 2001::/3 are currently being assigned by IANA for public use on IPv6 Internet. This is less than 2% of the total IPv6 address space.
• #5: Some of the special IPv6 addresses and prefixes are similar to IPv4. The ones shown here are particularly useful to remember. For example, IPv6 address ::/0 is analogous to IPv4 quad zero address 0.0.0.0 0.0.0.0, for use in defining static default routes. IPv6 address ::1/128 is analogous to IPv4 local host loopback address 127.0.0.1.
• #6: As with IPv4, Unicast is the largest and most important category of IPv6 addresses. The two most important address types in this category are Link-local and Global Routable addresses. Link-local addresses are used for communication between devices on a particular link. Global routable addresses are assigned to end devices for communication over the Internet. Each of these can be created either statically or dynamically as we will see later in this presentation. Note that every device or interface that has a Global routable address will also have a Link-local address.
• #7: Link-local addresses play an important role in the operation of IPv6. They are used for automatic address configuration, neighbor discovery, router discovery, and by many routing protocols. Packets with a link-local destination must stay on the link where they have been generated. Routers that could forward them to other links are not allowed to do so because there has been no verification of uniqueness outside the context of the origin link. Although link-local addresses are not routable they are often used by routing protocols for neighbor communication and updates.
• #8: When communicating using a link-local address, the outgoing interface must be specified because every IPv6 interface is connected to FE80::/10. For example, if you ping the neighbor’s link-local address, you will be asked to input the interface from which you wish to ping. Note that just enabling IPv6 on a router interface (Ethernet or Serial) and/or assigning the interface an IPv6 routable unicast address automatically creates an EUI link-local address. Link-local addresses can also be created statically using IOS commands.
• #9: The output from the show ipv6 interface command displays the Link-local address, which begins with FE80. The global unicast address, which begins with 2001, is also displayed. Global unicast addresses will be covered next.
• #10: A global unicast address is an IPv6 address from the global public unicast prefix (2001::/16). These addresses are routable on the global IPv6 Internet. The structure enables aggregation of routing prefixes to reduce the number of routing table entries in the global routing table. Global unicast addresses are aggregated upward through organizations and eventually to the ISPs. Additional information on IPv6 address subnetting and aggregation is provided later in this lesson.
• #11: The routable global unicast address typically consists of three parts: The first 48 bits of the address is the global routing prefix, which is assigned by an ISP and is derived from a /32 ISP prefix. The next 16 bits is the subnet ID (/49 thru /64), which may be use internally by an organization to subdivide its networks. The remaining 64 bits of the address is the device interface ID, typically in EUI-64 bit format, which will be discussed later.
• #12: The current IANA global routing prefix uses the range that starts with binary 0010 (2000::/3). Addresses with a prefix of 2000::/3 (0010) to E000::/3 (1110) are required to have 64-bit interface IDs in the extended universal identifier (EUI)-64 format. IPv6 prefix allocation is defined by policies at IANA and the regional registries.
• #13: The subnet ID can be used by an organization to create their own local addressing hierarchy. This 16-bit field allows up to 65,536 individual subnets. This number of subnets is comparable to an IPv4 network address such as 10.0.0.0/8 with 16 bits of subnetting using a /24 mask, however the IPv4 network/subnet combination will allow only 256 hosts per subnet and the IPv6 combination allows a huge number of hosts per subnet (2 to the 64th power).
• #14: In the sample output from the show ipv6 interface command, the /48 address assigned by the ISP is 2001:8:85A3, represented by three 16-bit fields. Note that the second 16-bit field in the output (with a value of Hex 0008) has the three leading zeros suppressed. The subnet number is Hex 4290 (which translates to the 17,040th subnet decimal) The interface address is 222:55FF:FE18:7DE, which is the EUI-64 host portion of the address. Also note that the /64 subnet is also clearly identified as highlighted in the output in green.
• #15: Multicasting is key to many IPv6 functions and performs a similar function as with IPv4. With IPv6 it is also a replacement for the broadcast address. IPv6 multicast addresses are defined by the prefix FF00::/8. An interface may belong to any number of multicast groups.
• #16: The first octet of a multicast address is always Hex FF or decimal 255 (eight binary ones). The second octet of the address contains the prefix and transient (lifetime) flags, and the scope of the multicast address.
• #17: The multicast addresses FF00:: to FF0F:: have the T flag set to 0 and are therefore permanent and reserved. For example, a multicast address starting with FF02::/16 is a permanent address. The last 4 bits of the /16 prefix represent the scope. For example, the scope of FF05 is site specific (Decimal 5 is binary 0101) and is used by Network Time Protocol (NTP) with IPv6.
• #18: Some examples of reserved IPv6 multicast address are listed here. IPv6 multicast address FF02::9 represents all RIP routers on a link. This is analogous to IPv4 multicast address 224.0.0.9 for RIPv2 routers. Address FF02::1:FFxx:xxxx is a solicited-node multicast addresses and is used for host autoconfiguration and neighbor discovery. It serves a function similar to ARP in IPv4.
• #19: With the solicited-node multicast addresses FF02::1:FFxx:xxxx, xx:xxxx is the far right 24 bits of the corresponding unicast or anycast address of the node. In the output shown, the value Hex 18:7DE8, which is the last 24 bits of the EUI-64 interface ID, is appended to the solicited-node multicast address (FF02::1:FF). Note that, when an IPv6 address is configured on the R1 interface, the interface automatically joins the “All Nodes” multicast group (FF02::1) and the “All Routers” multicast group (FF02::2).
• #20: The solicited-node multicast address (FF02::1:FF) is used for the neighbor discovery (ND) process and stateless address autoconfiguration. The Neighbor discovery (ND) process is a very important function in IPv6. It is used to determine the local-link address of a neighbor, the routers on the link and default route. It actively keeps track of neighbor reachability and sends network information from routers to hosts.
• #21: Neighbor Discovery (ND) uses four ICMPv6 packet types as described in the table: Neighbor Solicitation (NS) and Neighbor Advertisement (NA) ICMPv6 messages are used to determine link-layer addresses, neighbor reachability and for duplicate address detection (DAD). Router Solicitation (RS) and Router Advertisement (RA) ICMPv6 messages are used between hosts and routers to communicate prefixes and facilitate on-link determination or address configuration.
• #22: ICMPv6 Neighbor Solicitation (NS) is similar to IPv4 ARP in that it is used when resolving an IPv6 address to a MAC address. In the example, Host A needs to send a packet to Host B but needs the MAC address of host B. Host A sends a Neighbor Solicitation (ICMPv6 message type 135) on the link. The source address is the IPv6 unicast address of the source node, Host A.
• #23: Each destination node that receives the NS responds with an ICMPv6 NA message type 136. The source address of this message is the IPv6 address of the responding node, and the destination address is the IPv6 address of the original source node (which sent the NS). The data portion includes the link-layer address of the destination node (Host B). The two devices can now communicate on the link because they know each other’s link-layer addresses.
• #24: Stateless address autoconfiguration (SLAAC) offers new functionality in IPv6. Every IPv6 system is able to build its own unicast global address. This enables new devices such as cellular phones, wireless devices, home appliances, and home networks, to easily connect to the Internet. No configuration or DHCP server is required (other than the IPv6 address of the gateway router interface). An IPv6 router periodically sends network information on the local link using a Router Advertisement (RA). This includes the IPv6 prefix and default IPv6 route for use by hosts when autoconfiguring. IPv6 workstations and servers listen on the local link and configure themselves with an IPv6 Address and a default route. The IPv6 address is created by appending the host link-layer (MAC Address) in EUI-64 (Extended Unique Identifier) format to the /64 prefix sent by the router.
• #25: Shown here is a basic illustration how Autoconfiguration enables a host to configure its own IPv6 address (and default gateway) using a Router Advertisement (RA). The IPv6 host builds its IPv6 global unicast address by combining the prefix sent from the router and its Link-layer EUI-64 Address For stateless autoconfiguration to work, the router gateway interface IPv6 address must be pre-configured and unicast routing must be enabled on the interface. Note that when an IPv6 host is booting it sends out a Router Solicitation (RS) requesting routers to immediately generate an RA rather than wait for the next scheduled RA.
• #26: With EUI-64 IPv6 addresses the first 64 bits are the network portion of the address and are either specified statically or learned via stateless autoconfiguration. The interface ID (second 64-bits) is the host portion of the address and is automatically generated by the router or host device. The interface ID on an Ethernet link is based on the 48-bit MAC address of the interface with an additional 16-bit 0xFFFE inserted in the middle of the MAC address. This creates an extended unique identifier referred to as the EUI-64 format. EUI-64 addresses can be generated on IPv6 router interfaces and on IPv6 hosts.
• #27: The EUI-64 standard inserts a 16-bit Hex value FFFE in the middle at the 24th bit of the MAC address to create a unique 64-bit interface identifier. The seventh bit in the high-order byte is set to 1 to indicate the uniqueness of the interface ID.
• #28: This graphic provides some additional information on how the Host/Router stateless autoconfiguration process works. When Host A boots up, it sends a router solicitation (RS) message to the “All Routers” multicast address (FF02::2) requesting a router advertisement (RA). Router R1 on the link responds with an RA to the “All Nodes” multicast address (FF02::1) so that Host A has the information it needs to autoconfigure its IPv6 address. The prefix included in the RA is used as the /64 prefix for the host address. The interface ID used is the EUI-64 format interface ID for the host.
• #29: The host now creates an IPv6 address using the RA supplied by the router, as well as a link-local address and a solicited-node address. Next it needs to verify that it’s new IPv6 address is unique on the link using the Duplicate Address Detection (DAD) process. DAD is used during the autoconfiguration process to ensure that no other device is using the autoconfiguration address. During the DAD phase, Host A sends an NS to query if another node on the link has the same IPv6 address. If a node responds to the request, it means that the IPv6 address is already in use, and Host A needs to be manually configured.
• #30: This section provides basic information on IPv6 subnetting and address aggregation.
• #31: IPv6 subnetting is based on the network address assigned by the ISP and the needs of the organization to subdivide that space. In this example, The network address assigned to the organization is 2001:0025:0012 (the first 48 bits). The leading zeros are suppressed in the address. The site prefix is /48 decimal and the site prefix plus subnets (AB12) is /64. IPv6 uses the same notation as CIDR in IPv4. There are no subnet masks, only a slash (/) and a decimal number representing the number of bits in the network/subnet. In addition, there are no reserved subnet addresses (network or broadcast) with IPv6 so it is not necessary to account for the first and last address on a subnet when determining the number of valid host addresses (no more n^2 – 2).
• #32: IPv6 uses CIDR notation, as does IPv4, to identify the prefix length. The normal prefix length (total number of network bits) for an IPv6 address is 64. The rightmost 16 bits of the prefix (/49 to /64) allow for 65,535 LANs.
• #33: The global routing prefix (/32) is assigned to a service provider by the Internet Assigned Numbers Authority (IANA). The next 16 bits is the SLA or site level aggregator and is a /48 network address assigned to a customer by their service provider. The SLA can be viewed as a subnet ID from the ISPs perspective. The 16-bit LAN ID represents individual networks within the customer site and is administered by the customer. The total network ID is comprised of the Global prefix (32 bits), The SLA (16 bits) and the LAN subnet ID (16 bits) resulting in a prefix length of /64. The remaining 64 bits are the Interface ID and represent an individual host on a subnet or LAN.
• #34: This slide summarizes the prefix lengths and subnet/host structure of a standard IPv6 global unicast address.
• #35: The first three bits are fixed (001) which results in 61 Global Network bits and 64 Host bits. You do NOT get to “borrow” host bits any more (nor is it necessary, due to the huge IPv6 address space) 2^64 hosts are theoretically possible in a single broadcast domain. With autoconfiguration it is not necessary to configure them. Host devices can be moved around and will automatically generate their EUI-64 addresses. VLANs become the method of isolation.
• #36: The large IPv6 address space is hierarchical by design which allows for multiple levels, greatly facilitating address assignment and aggregation.
• #37: The IPV6 address space is huge and hierarchical. A prefix can be assigned to an organization that can handle the entire network, even for very large organizations, and still have plenty of room for future growth. ISPs summarize routes so that all customer prefixes can be aggregated into one prefix and made available to the Internet. Aggregation provides efficient and scalable routing and fewer routes in the global IPV6 routing table with an enforced hierarchical VLSM model.
• #38: In this example, all subnets of the /48 networks assigned to Customers A1 and A2 can be summarized by ISP A to a single /35 network address to be advertised to the Internet. Note that if the customer changes their provider, they must change the IPv6 Prefix.
• #39: In this example, an ISP has been assigned a /32 address by the regional registry. This leaves 16 SLA bits which yields 65,535 /48 networks that can be assigned to various customer sites.
• #40: The customer can subdivide the 16 bits as desired to number their internal /64 subnets. Several examples are shown here. For a smaller organization, the ISP might assign a /56 prefix instead of a /64, which would still allow the customer to create 256 subnets (2^8 = 256).
• #41: Applying one of these /64 subnet prefixes to a router interface allows the router to provide it to hosts on the subnet link so they can autoconfigure their IPv6 addresses.
• #42: This section provides an examples IPv6-related commands that can be used to configure and verify IPv6 implementations.
• #43: Many IPv6 show commands are available to verify IPv6 implementations. Most of them are similar to their IPv4 counterparts with the only difference being the replacement of the keyword “ip” with “ipv6” in the command. For example, the IPv6 version of the IPv4 show ip interface command is show ipv6 interface.. The IPv4 show ip route command becomes show ipv6 route. Some useful IPv6 show commands are listed here. Note that the commands show ipv6 routers and show ipv6 neighbors are unique to IPv6.
• #44: Several useful IPv6 debug commands are also available to verify IPv6 implementations.
• #45: The Cisco IOS ipv6 unicast routing command is required before implementing either static or dynamic routing with IPv6 global unicast addresses. The command is not needed before configuring IPv6 interface addresses. It is also required for the interface to provide stateless auto-configuration. Configuring no ipv6 unicast-routing disables the IPv6 routing capabilities of the router and the router acts as an IPv6 end-station.
• #46: The command ipv6 cef enables express forwarding and improves overall packet processing performance. It is disabled by default.
• #47: Link-local and global unicast addresses can be assigned statically or dynamically. This section covers manual assignment of these addresses.
• #48: To configure a global or link-local address on a router interface use the ipv6 address command. You may manually configure a specific global or link-local unicast address or allow the IOS to generate it by specifying eui-64. The link-local parameter configures the address as the link-local address on the interface. The omission of the link-local parameter configures the address as the global unicast IPv6 address on the interface. The eui-64 parameter completes a global IPv6 address automatically using an EUI-64 format interface ID.
• #49: Link-local addresses are created: Automatically using the EUI-64 format if the interface has IPv6 enabled on it or a global IPv6 address configured. Manually by specifying an interface ID. Manually configured interface IDs are easier to remember than EUI-64 generated IDs. Note that the prefix mask is not required on link-local addresses because they are not routed. This example configures a manual Link-local address of FE80::1 (actually FE80:0:0:0:0:0:0:1).
• #50: The output from the show ipv6 interface command confirms the link-local address previously configured and that no IPv6 global unicast is configured yet.
• #51: Global Unicast IPv6 addresses are assigned manually by omitting the link-local parameter. For example, IPv6 address 2001:1::1/64 is configured on R1’s Fast Ethernet 0/0 interface. The entire address is manually configured and the EUI-64 format was not used. Note that, if an IPv6 global address is configured first or if ipv6 is simply enabled on the router interface, an EUI-64 Link-local address (FE80) will automatically be created.
• #52: The output from the show ipv6 interface command confirms that a static IPv6 global unicast address is now configured on the F0/0 interface.
• #53: In this IPv6 global unicast example, a 64-bit prefix is assigned and the Link-local interface ID is generated automatically using EUI-64. Configuring a global unicast IPv6 address on an interface also automatically generates a link-local interface (EUI-64) address.
• #54: In this example, a /64 prefix is provided and the interface ID is generated automatically using EUI-64.
• #57: In this section, we configure an IPv6 global unicast address on an interface statically using the address of another interface.
• #58: This feature is similar to its IPv4 counterpart. It enables IPv6 processing on an interface without assigning an explicit IPv6 address to the interface. The unnumbered interface will use the IPv6 address of the interface specified by the interface-type interface-number parameters as the source address of traffic from the configured interface. The interface specified in the command must be in the “up” state.
• #59: IPv6 supports unnumbered interfaces to enable IPv6 processing on an interface without assigning an explicit IPv6 address to the interface. In this example, a loopback interface is created and configured with an IPv6 address. The Serial 0/0/0 interface is then configured to use the IPv6 address of the loopback interface.
• #60: The output confirms that the Serial 0/0/0 interface uses the IPv6 address from interface loopback 10.
• #61: In this section, we configure a router interface to support stateless autoconfiguration of hosts on the local link.
• #62: In this example, the router interface can autoconfigure itself by appending its IPv6 interface identifier (in EUI-64 format) to the local link 64-bit prefix received from another router on the link. (Optional) If the default keyword is used it causes a default route to be installed using that default router. The keyword can be specified only on one interface.
• #63: This feature enables a router to detect unavailable neighbors more quickly. The milliseconds parameter (from 0 to 3,600,000) configures the amount of time that a neighbor waits before sending an update to the router. Default is 0 milliseconds (unspecified time) in router advertisements and 30,000 (30 seconds) for the neighbor discovery activity. Caution: A very short time may consume more network bandwidth and processing resources.
• #64: This command statically configures an entry in the IPv6 neighbor discovery cache, mapping the IPv6 address to the hardware address on an interface.
• #65: A router interface IPv6 address can be configured using the autoconfig option (if another router on the link that can provide the prefix) or by specifying the /64 prefix and letting the router autoconfigure its IP address (as shown here) using EUI-64. When configuring the interface, only the network portion is supplied in the command. Autoconfiguration is normally used on interfaces that have MAC addresses. The ipv6 enable command automatically configures an IPv6 link-local unicast address on the interface while also enabling the interface for IPv6 processing.
• #66: The router automatically generates its link-local address (FE80). The globally routable address is generated by stateless autoconfiguration (2001). Note that the show command includes “ipv6”