You charge apperceive IPv4 acclamation carefully to alike ability this point in your CCIE abstraction efforts.
This affiliate takes advantage of that actuality to advice you bigger apprentice about IPv6 acclamation by making
comparisons amid IPv4 and IPv6. But first, you charge to briefly analyze why we charge IPv6 or,
more absolutely perhaps, why we will charge it in the future.
IPv6 was created to accommodated the charge for added host addresses than IPv4 can accommodate—a lot
more. In the aboriginal 1990s, back the cardinal of Internet-connected hosts began to appearance signs of
massive growth, article of a crisis was brewing amid the standards bodies about how to deal
with that advance in a way that would calibration not aloof to the concise need, but continued appellation as well.
It takes a lot of assay and time to actualize a new acclamation accepted that meets those goals.
Internet advance appropriate faster solutions than a absolute new acclamation accepted could support.
Two methods were bound implemented to accommodated the concise need: RFC 1918 clandestine IP
addresses and NAT/PAT. In a way, these techniques accept been so acknowledged at abridgement the growth
of Internet acquisition tables that they accept pushed out the charge for IPv6 by at atomic a decade, but that
need still exists. The day is advancing back the apple will artlessly accept to move to IPv6 for reasons
of appliance requirements, if not for near-term burnout of IPv4 addresses. One disciplinarian in this
progression is peer-to-peer applications, which accept developed abundantly in acceptance and are complex
to abutment with NAT/PAT. Addition is that the amoebic advance of the Internet about IPv4 has led
to suboptimal and bare abode allocation amid the busy areas of the world,
especially because the billow in Internet advance in awful busy countries that were not part
of the aboriginal Internet explosion.
IPv6 gives us a adventitious to admeasure abode ranges in a added alive way, which will ultimately
optimize Internet acquisition tables. At the aforementioned time, IPv6 provides an about unimaginably vast
pool of host IP addresses. At some point, NAT may become a abroad anamnesis of an ancient age.
Let’s appraise what makes IPv6 what it is and how it differs from IPv4. The key differences in
IPv6 acclamation compared to IPv4 follows:
■ IPv6 addresses are 128 $.25 long, compared to 32 $.25 continued for IPv4. In added words, IPv6
addresses are 296 times added abundant than IPv4 addresses.
■ IPv6 addresses are represented in hexadecimal rather than decimal and use colon-separated
fields of 16 $.25 each, rather than decimal credibility amid 8-bit fields, as in IPv4.
746 Affiliate 20: IP Version 6
■ In a Cisco IOS router, you can configure assorted IPv6 addresses on an interface (logical or
physical), all of them with according antecedence in agreement of the interface’s behavior. By
comparison, you can configure alone one primary IPv4 abode per interface with optional
secondary addresses.
■ Globally altered IPv6 addresses can be configured automatically by a router appliance the builtin
autoconfiguration action afterwards the abetment of protocols such as DHCP.
■ IPv6 uses congenital acquaintance discovery, by which an IPv6 bulge can ascertain its neighbors and
any IPv6 routers on a segment, as able-bodied as whether any routers present are accommodating to serve as
a absence aperture for hosts.
■ The concepts of clandestine IPv4 acclamation in RFC 1918 do not administer to IPv6; however, several
different types of IPv6 addresses abide to accommodate agnate functionality.
The above-mentioned account provides several key differences amid IPv4 and IPv6; the abutting section
explores the capacity of these concepts and provides an accession to IPv6 agreement in
Cisco IOS.
IPv6 Acclamation and Abode Types
This area covers the basics of IPv6 addressing, starting with how IPv6 addresses are
represented and again exploring the altered types of IPv6 addresses. Afterwards laying that foundation,
the “Basic IPv6 Functionality Protocols” area gets into the ancestors of protocols that enables
IPv6 to absolutely action as a arrangement band protocol.
IPv6 Abode Notation
Because of the breadth of IPv6 addresses, it is abstract to represent them the aforementioned way as IPv4
addresses. At 128 bits, IPv6 addresses are four times the breadth of IPv4 addresses, so a more
efficient way of apery them is alleged for. As a result, anniversary of the eight groups of 16 $.25 in
an IPv6 abode is represented in hex, and these groups are afar by colons, as follows:
1234:5678:9ACB:DEF0:1234:5678:9ABC:DEF0
In IPv6, as in IPv4, unicast addresses accept a two-level network:host bureaucracy (known in IPv6 as
the prefix and interface ID) that can be afar into these two genitalia on any bit abuttals in the
address. The prefix allocation of the abode includes a brace of components, including a global
routing prefix and a subnet. However, the two-level bureaucracy separates the prefix from the
interface ID abundant like it divides the arrangement and host portions of an IPv4 address. Instead of using
a decimal or hex subnet mask, though, IPv6 subnets use carve characters to announce the network
portion of the address, as follows:
1234:5678:9ABC:DEF0:1234:5678:9ABC:DEF0/64
IPv6 Acclamation and Abode Types 747
An IPv6 abode with a prefix breadth of 64 bits, frequently alleged a /64 abode in this context, sets
aside the aboriginal bisected of the abode amplitude for the prefix and the aftermost bisected for the interface ID. After
more advantage of the arena rules for IPv6 addressing, this affiliate covers the agency that prefixes
and interface IDs are developed for unicast addresses, as able-bodied as the added abode types used
in IPv6 networks.
Address Abridgement Rules
Even in the almost able architecture apparent earlier, the antecedent IPv6 addresses can be
cumbersome because of their arduous length. As a result, a brace of abridgement methods are used
to accomplish it easier for us to assignment with them. These methods accommodate the following:
■ Whenever one or added alternating 16-bit groups in an IPv6 abode abide of all 0s, that
portion of the abode can be bare and represented by two colons (::). The two-colon
abbreviation can be acclimated alone already in an address, to annihilate ambiguity.
■ Back a 16-bit accumulation in an IPv6 abode begins with one or added 0s, the arch 0s can be
omitted. This advantage applies behindhand of whether the double-colon abridgement adjustment is
used anywhere in the address.
Here are some examples of the above-mentioned techniques, accustomed an IPv6 abode of
2001:0001:0000:0000:00A1:0CC0:01AB:397A. Accurate agency of abridgement this abode appliance the
preceding rules accommodate these:
2001:1:0:0:A1:CC0:1AB:397A
2001:0001::00A1:0CC0:0174AB:397A
2001:1::A1:CC0:1AB:397A
All of these abbreviated examples actually represent the accustomed abode and can be
independently interpreted by any IPv6 host as the aforementioned address.
IPv6 Abode Types
Like IPv4 addresses, several types of IPv6 addresses are appropriate for the assorted applications of
IPv6 as a Band 3 protocol. In IPv4, the abode types are unicast, multicast, and broadcast. IPv6
differs hardly in that advertisement acclamation is not used; appropriate multicast addresses booty the place
of IPv4 advertisement addresses. However, three abode types abide in IPv6: unicast, multicast, and
anycast. This area of the affiliate discusses anniversary one. Table 20-2 summarizes the IPv6 address
types.
748 Affiliate 20: IP Version 6
Many of the agreement in Table 20-2 are absolute to IPv6. The afterward sections appraise anniversary of the
address types listed in the table.
Unicast
Unicast IPv6 addresses accept abundant the aforementioned functionality as unicast IPv4 addresses, but because
IPv6’s 128-bit abode amplitude provides so abounding added addresses to use, we accept abundant more
flexibility in allotment them globally. Because one of the intents for IPv6 acclamation in public
networks is to acquiesce advanced use of globally altered addresses, aggregatable all-around unicast IPv6
addresses are allocated in a way in which they can be calmly abbreviated to analytic accommodate the
size of all-around IPv6 acquisition tables in account provider networks.
In accession to aggregatable all-around unicast addresses, several added aspects of IPv6 unicast
addressing deserve acknowledgment actuality and chase in the abutting few sections.
Aggregatable All-around Addresses
In accepted usage, aggregatable all-around addresses are assigned from the IPv6 addresses that begin
with bifold 001. This amount can be accounting in prefix characters as 2000::/3, which agency “all IPv6
addresses whose aboriginal 3 $.25 are according to the aboriginal 3 $.25 of hex 2000." In practice, this includes IPv6
addresses that activate with hex 2 or 3. (Note that RFC 3587 afterwards removed the brake to only
allocate aggregatable all-around unicast addresses from the 2000::/3, but in practice, these addresses
are still allocated from this range.) To ensure that IPv6 addresses can be abbreviated efficiently
when advertised against Internet routers, several all-around organizations admeasure these addresses to
service providers and added users. See RFC 3587 and RFC 3177 for added details.
Table 20-2 IPv6 Abode Types
Address Blazon Ambit Application
Aggregatable global
unicast
2000::/3 Host-to-host communication; aforementioned as IPv4
unicast.
Multicast FF00::/8 One-to-many and many-to-many
communication; aforementioned as IPv4 multicast.
Anycast Aforementioned as Unicast Application-based, including amount balancing,
optimizing cartage for a accurate service, and
redundancy. Relies on acquisition metrics to
determine the best destination for a particular
host.
Link-local unicast FE80::/10 Connected-link communications.
Solicited-node multicast FF02::1:FF00:0/104 Acquaintance solicitation.
IPv6 Acclamation and Abode Types 749
Aggregatable all-around abode prefixes are structured so that they can be carefully abbreviated and
aggregated through a bureaucracy consisting of a clandestine arrangement and a alternation of account providers.
Here is how that works, based on RFC 3177, starting afterwards the aboriginal 3 $.25 in the prefix:
■ The abutting 45 $.25 represent the all-around acquisition prefix.
■ The aftermost 16 $.25 in the prefix, anon above-mentioned the Interface ID allocation of the address,
are Site Level Aggregator (SLA), bits. These $.25 are acclimated by an alignment for its own
internal acclamation hierarchy. This acreage is additionally accepted as the Subnet ID.
■ The aftermost 64 $.25 accomplish up the interface ID.
Figure 20-1 shows the aggregatable all-around unicast IPv6 abode format.
Figure 20-1 IPv6 Abode Format
The interface ID allocation of an aggregatable all-around IPv6 abode can be absolutely assigned in Cisco
IOS or acquired appliance a cardinal of methods explored afterwards in this affiliate in the “IPv6 Address
Autoconfiguration” section. These addresses should use an Interface ID in the adapted EUI-64
format, discussed afterwards in this chapter. Depending on how these addresses are assigned, however,
the Universal/Local bit, which is the 7th bit in the Interface ID acreage of an IPv6 address, can be set
to 0 (locally administered) or 1 (globally unique) to announce the attributes of the Interface ID portion
of the address.
Link-Local Addresses
As the appellation implies, link-local addresses are acclimated on a abstracts articulation or multiaccess network, such as
a consecutive articulation or an Ethernet network. Because these addresses are link-local in scope, they are
guaranteed to be altered alone on that articulation or multiaccess network. Anniversary interface type, regardless
of whether it is serial, PPP, ATM, Frame Relay, Ethernet, or article else, gets a link-local
address back IPv6 is enabled on that interface.
64 Bits
16 Bits
SLA or
Subnet ID
45 Bits
48 Bits
Global Prefix
3 Bits
001
Interface
Prefix ID
750 Affiliate 20: IP Version 6
Link-local addresses consistently activate with FE80::/10. The Interface ID allocation of the abode is
derived appliance the adapted EUI-64 format, discussed afterwards in this chapter. The actual 54 bits
of the prefix are consistently set to 0.
On Ethernet interfaces, the IEEE 802 MAC abode is the base for the Interface ID. For other
interface types, routers draw from a basin of basic MAC addresses to accomplish the Interface IDs.
An archetype of a absolutely formed link-local abode follows:
FE80::207:85FF:FE80:71B8
As you ability accumulate from the name, link-local addresses are acclimated for advice between
hosts that do not charge to leave the bounded segment. By definition, routers do not advanced link-local
traffic to added segments. As you will see afterwards in this chapter, link-local addresses are acclimated for
operations such as acquisition agreement acquaintance communications, which are by their attributes linklocal.
IPv4-Compatible IPv6 Addresses
Many alteration strategies accept been developed for IPv4 networks to drift to IPv6 account and
for IPv6 networks to intercommunicate over IPv4 networks. Most of these strategies involve
tunneling. Similarly, a apparatus exists for creating IPv6 addresses that are accordant with
IPv4. These addresses use 0s in the aboriginal 96 $.25 of the abode and one of the two formats for the
remaining allocation of the address. Booty a attending at an example, accustomed the IPv4 abode 10.10.100.16.
The afterward are accurate IPv4-compatible IPv6 addresses that accord to this IPv4 abode (all
of these are in hexadecimal, as IPv6 addresses are universally represented):
0:0:0:0:0:10:10:100:16
::10:10:100:16
::A:A:64:10
IPv4-compatible IPv6 addresses are not broadly acclimated and do not represent a architecture best practice,
but you should be accustomed with their format. See the area “Tunneling,” afterwards in this affiliate for
more detail on IPv4-compatible abode acceptance in the agnate adit blazon and on the
deprecation of this tunneling blazon in Cisco IOS.
Assigning an IPv6 Unicast Abode to a Router Interface
To configure any IPv6 abode or added IPv6 feature, you charge aboriginal globally accredit IPv6 on the
router or switch:
Stengel(config)# ipv6 unicast-routing
IPv6 Acclamation and Abode Types 751
Next, configure a all-around unicast address:
Stengel(config-if)# ipv6 abode 2001:128:ab2e:1a::1/64
Routers automatically configure a articulation bounded IPv6 abode on all IPv6-enabled interfaces. However,
you can configure the articulation bounded abode with the afterward command. (Note the the link-local
keyword to baptize the abode type.)
Stengel(config-if)# ipv6 abode fe80::1 link-local
Unlike IPv4, IPv6 allows you to accredit abounding addresses to an interface. All IPv6 addresses
configured on an interface get according antecedence in agreement of IP acquisition behavior.
Multicast
Multicast for IPv6 functions abundant like IPv4 multicast. It allows assorted hosts to become
members of (that is, accept cartage beatific to) a multicast accumulation afterwards attention to their area or
number. A multicast receiver is accepted as a accumulation member, because it joins the multicast accumulation to
receive traffic. Multicast addresses in IPv6 accept a specific format, which is covered in the next
section.
Because IPv6 has no advertisement acclamation concept, multicast takes the abode of all functions that
would use advertisement in an IPv4 network. For example, the IPv6 DHCP action uses multicast for
sending cartage to an alien host on a bounded network.
As in IPv4, IPv6 multicast addresses are consistently destinations; a multicast abode cannot be used
as a antecedent of any IPv6 traffic.
IPv6 multicast is covered in added detail in the aftermost area of this chapter.
IPv6 Multicast Abode Format
Multicast addresses in IPv6 consistently activate with FF as the aboriginal octet in the address, or FF00::/8. The
second octet specifies the lifetime and ambit of the multicast group. Lifetime can be abiding or
temporary. Ambit can be bounded to any of the following:
■ Node
■ Link
■ Site
■ Organization
■ Global
752 Affiliate 20: IP Version 6
The multicast abode architecture is apparent in Figure 20-2.
Figure 20-2 IPv6 Multicast Abode Format
Table 20-3 shows several acclaimed IPv6 multicast accumulation addresses and their functions.
In an IPv6 network, as in IPv4, there is an all-nodes multicast accumulation (FF02::1), of which all IPv6
hosts are members. All routers charge accompany the all-routers multicast abode (FF02::2). In addition,
IPv6 multicast uses a solicited-node accumulation that anniversary router charge accompany for all of its unicast and
anycast addresses. The architecture for solicited-node multicast addresses is
FF02::1:FF00:0000/104
Table 20-3 IPv6 Multicast Well-Known Addresses
Function Multicast Accumulation IPv4 Equivalent
All hosts FF02::1 Subnet advertisement address
All Routers FF02::2 224.0.0.2
OSPFv3 routers FF02::5 224.0.0.5
OSPFv3 appointed routers FF02::6 224.0.0.6
EIGRP routers FF02::A 224.0.0.10
PIM routers FF02::D 224.0.0.13
Interface ID
64 Bits
All 0s
8 Bits
FF
8 Bits
Lifetime Scope
4 $.25 4 Bits
16 Bits
0000 = Permanent
0001 = Temporary
0001 = Node
0010 = Link
0101 = Site
1000 = Organization
1110 = Global
Scope
IPv6 Acclamation and Abode Types 753
Note that all but the aftermost 24 $.25 of the abode are defined by the /104 prefix. Solicited-node
addresses are congenital from this prefix concatenated with the low-order 24 $.25 (128 – 104 = 24) of
the agnate unicast or anycast address. For example, a unicast abode of
2001:1AB:2003:1::CBAC:DF01
has a agnate solicited-node multicast abode of
FF02::1:FFAC:DF01
Solicited-node addresses are acclimated in the Acquaintance Analysis (ND) process, covered afterwards in this
chapter.
Multicast in IPv6 relies on a cardinal of protocols with which you are already familiar, including
PIM. Multicast Listener Analysis is addition key allotment of IPv6 multicast. These capacity and other
related multicast capacity are covered afterwards in this affiliate in the “IPv6 Multicast” section.
Anycast
In some applications, decidedly server farms or provider environments, it may be adorable to
pool a cardinal of servers to accommodate redundancy, amount balancing, or both. Several protocols can
provide this functionality in IPv4 networks.
IPv6 has congenital abutment for this appliance in the anatomy of anycast addressing. Anycast addresses
can be assigned to any cardinal of hosts that accommodate the aforementioned service; back added hosts admission this
service, the specific server they hit is bent by the unicast acquisition metrics on the aisle to that
particular accumulation of servers. This provides geographic differentiation, added availability, and
load acclimation for the service.
Anycast addresses are fatigued from the IPv6 unicast abode basin and, therefore, are not
distinguishable from unicast addresses. RFC 2526 recommends a ambit of addresses for use by
anycast applications. Already an abode is assigned to added than one host, it becomes an anycast
address by definition. Because anycast addresses cannot be acclimated to antecedent traffic, however, a
router charge apperceive if one of its interface IPv6 addresses is an anycast address. Therefore, Cisco IOS
Software requires the anycast keyword to be activated back an anycast abode is configured, as in
this example:
Mariano(config-if)# ipv6 abode 3001:fffe::104/64 anycast
All IPv6 routers additionally charge abutment the subnet router anycast address. This anycast address
is a prefix followed by all 0s in the interface ID allocation of the address. Hosts can use a subnet
router anycast abode to ability a accurate router on the articulation articular by the prefix accustomed in the
subnet router anycast address.
The Bearding Address
One added blazon of IPv6 abode deserves acknowledgment in this section, as it is acclimated for a cardinal of
functions in IPv6 communications. This address, which is acclimated for some types of requests covered
later in this chapter, is represented artlessly by ::. The bearding abode is consistently a antecedent address
used by an interface that has not yet abstruse its unicast address. The bearding abode cannot be
assigned to an interface, and it cannot be acclimated as a destination address.