Offset ZeroCCNA 200-301 study captureGet the app
Menu

IPv6Area 1C · IPv6 addressing · Weight (our estimate) ≈3%

IPv6 addressing for the CCNA: read the prefix, name the type, build the interface ID

Every IPv6 address you meet on the CCNA decodes in three reads: the leading bits name the address type, the prefix length marks the network, and the rest is an interface ID you either configure or derive from the MAC address with modified EUI-64. Compression rules sit on top, since addresses are usually written in their short form.

Reviewed October 29, 2026

Area1C
Weight (our estimate)≈3%
Frames15
Frames · 8 sections

Address types by their leading bits

Global unicast, 2000::/3
Routable on the internet. A provider assigns a prefix, the site carves subnets from it, and each host fills in its interface ID (RFC 4291).
Unique local, fc00::/7
Internal addressing that the internet does not route. In practice these addresses start with fd, because the eighth bit marks a locally assigned prefix (RFC 4193).
Link-local, fe80::/10
Scoped to a single link. Neighbor discovery and routing protocols source their messages from it, which is why IPv6 routes learned from a neighbor often show a link-local next hop.
Multicast, ff00::/8
One-to-many delivery. IPv6 has no broadcast; groups such as ff02::1 (all nodes on the link) and ff02::2 (all routers) do the jobs broadcast used to do.
Anycast
No prefix of its own: an ordinary unicast address assigned to more than one node, and the routing system hands each packet to just one of those nodes.
Loopback ::1 and unspecified ::
::1 is the host talking to itself. :: means no address yet, as in the source of a packet sent before the host has one.

IPv4 and IPv6, field by field

Where the two protocols differ in ways the stems rely on
Point of comparisonIPv4IPv6
Address length32 bits128 bits
Written asFour decimal octetsEight hex hextets, compressible
BroadcastYes, per subnetNone; multicast takes its jobs
Address resolutionARPNeighbor Discovery (ICMPv6)
Host addressingStatic or DHCPStatic, SLAAC or DHCPv6
Basic header20 to 60 bytes, with a checksumFixed 40 bytes, no checksum
FragmentationSender or routerSender only

Sources: RFC 8200 for the header and fragmentation, RFC 4861 for Neighbor Discovery.

Modified EUI-64, one MAC address through it

0000MAC0050.56c0.0a1b0010Split in half0020Insert FFFE0030Flip bit 70040Interface ID
Figure 1.MAC 0050.56c0.0a1b becomes the interface ID 0250:56ff:fec0:0a1b.
  • 0000 48 bits from the interface
  • 0010 0050.56 | c0.0a1b
  • 0020 0050:56ff:fec0:0a1b
  • 0030 first byte 00 becomes 02
  • 0040 0250:56ff:fec0:0a1b, after the /64 prefix

Writing it short, reading it long

Two rules compress an address, and RFC 5952 adds the preferred form. The exhibit below runs one address through them.

  1. Drop leading zeros in each hextet: 0db8 becomes db8 and 0e10 becomes e10. Trailing zeros stay, so e10 never shrinks to e1.
  2. Collapse a single stretch of consecutive 0000 hextets into ::. Only one, because a reader could not tell how many zeros each of two :: stood for.
  3. Use :: for the longest run, write a lone zero hextet as 0, and keep hex in lowercase. The result: 2001:db8:0:42::e10.

To expand, count the hextets that are present and let :: stand for the rest of the eight.

One address, compressed in two steps

full     2001:0db8:0000:0042:0000:0000:0000:0e10rule 1   2001:db8:0:42:0:0:0:e10rule 2   2001:db8:0:42::e10

Prefixes and the router

Prefix lengths read the same way as in IPv4. A site prefix such as 2001:db8:acad::/48 leaves 16 bits of subnet ID before the /64 boundary that LAN subnets use, which is 65,536 subnets for one site.

IOS needs ipv6 unicast-routing before it forwards IPv6 between interfaces; without it the interfaces still take addresses, and the router behaves like a host. Hosts can build their own addresses with stateless address autoconfiguration (SLAAC, RFC 4862): the router advertises the prefix and the host supplies the interface ID.

Three ways to address one interface (lab values)

R2(config)#ipv6 unicast-routingR2(config)#interface GigabitEthernet0/1R2(config-if)#ipv6 address 2001:db8:acad:5::/64 eui-64R2(config-if)#ipv6 address 2001:db8:acad:6::1/64R2(config-if)#ipv6 address fe80::2 link-local

With MAC 0050.56c0.0a1b on Gi0/1, the eui-64 line produces an address ending in 250:56ff:fec0:a1b. show ipv6 interface brief lists all three addresses.

Frames: IPv6 addressing

Prefix reading, type recognition, compression and EUI-64. Open the notes after each pick: they show which bit or hextet decided it.

Practice frames written by our editors for study, not Cisco items. The 200-301 exam itself mixes multiple choice, drag-and-drop and lab-style items.

Frame 001 of 015 · 0 answered · 0 ok

Keys

1–5 or A–E pick an option (on choose-N frames they toggle; then Check answer) · ← → previous / next frame · ↑ ↓ move in the frame list

  1. Frame 001 of 015

    What is the primary reason for using unicast addressing in IPv6, and how does it function differently from anycast addressing?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      One-to-many delivery is multicast, not unicast, and anycast is not a form of multicasting, since only one receiver gets each packet.

    2. Note on B

      Correct: a unicast address identifies one interface, while an anycast address is shared by several interfaces and routing delivers to the nearest one.

    3. Note on C

      How an address is assigned (SLAAC, DHCPv6 or by hand) does not define its type; unicast and anycast can both be configured manually.

    4. Note on D

      Unicast includes global unicast, which is routed worldwide, as well as link-local; anycast comes from the unicast space and needs no special routing.

    Best answer: B · why

    A unicast address identifies exactly one interface, so a packet sent to it reaches that interface alone. An anycast address is a unicast address assigned to several interfaces, usually on different devices, and routing delivers each packet to the nearest one by routing metric. One-to-many delivery is multicast, which is a separate address type.

  2. Frame 002 of 015

    What is the size of the Interface Identifier in an IPv6 address?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      16 bytes is 128 bits, the length of the whole IPv6 address, not just the interface ID.

    2. Note on B

      32 bits is the length of an entire IPv4 address; the IPv6 interface ID is twice that.

    3. Note on C

      Correct: unicast IPv6 addresses normally split into a 64-bit prefix and a 64-bit interface identifier.

    4. Note on D

      4 bytes is 32 bits, half the size of the interface identifier.

    Best answer: C · why

    An IPv6 address is 128 bits. In a typical unicast address the first 64 bits are the network prefix (global routing prefix plus subnet ID) and the last 64 bits are the interface identifier, which is set manually, generated randomly or built with modified EUI-64.

  3. Frame 003 of 015

    In a computer network, which type of IP address is analogous to a private IPv4 address in a local network environment?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      Link-local addresses (FE80::/10) work only on one link and are never routed, so they are narrower than a private LAN range.

    2. Note on B

      Global unicast addresses (2000::/3) are routable on the internet, which makes them the counterpart of public IPv4 addresses.

    3. Note on C

      Correct: unique local addresses (FC00::/7, in practice FD00::/8) route inside an organization but not on the internet, like RFC 1918 space.

    Best answer: C · why

    Unique local addresses, FC00::/7 (used in practice as FD00::/8), are meant for routing inside a site or organization and are not routed on the public internet, the role that private RFC 1918 space plays in IPv4. Global unicast addresses match public IPv4 addresses, and link-local addresses work only on a single link.

  4. Frame 004 of 015

    In an IPv6 routing table, the default route is represented as ::/0. How does the router handle a packet destined to a link-local address if no specific route exists?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      No next hop applies: a link-local destination is valid only on its own link, so the router does not route it anywhere.

    2. Note on B

      Routers never rewrite a destination from link-local to global; IPv6 has no such translation step.

    3. Note on C

      Correct: link-local addresses (FE80::/10) are valid only on one link, so a router never forwards a packet with a link-local destination to another link.

    4. Note on D

      The default route applies only to routable destinations; a link-local destination is not forwarded off its link even with ::/0 in the table.

    Best answer: C · why

    Link-local addresses are only used on the local network segment and are not forwarded by routers. Therefore, a packet destined for a link-local address is never routed beyond its local link.

  5. Frame 005 of 015

    What is the prefix for the following IPv6 address: 2001:0db8:85a3::8a2e:0370:7334 / 48?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      This is the end of the address, the interface part; a prefix is always taken from the leftmost bits.

    2. Note on B

      2001:0db8::/48 keeps only two hextets, 32 bits; a /48 needs the first three hextets.

    3. Note on C

      Correct: /48 means the first 48 bits, three 16-bit hextets, 2001:0db8:85a3, with the rest set to zero: 2001:0db8:85a3::/48.

    4. Note on D

      A prefix comes from the leftmost bits; these hextets are from the right half of the address.

    Best answer: C · why

    Each IPv6 hextet is 16 bits, so a /48 prefix is the first three hextets: 2001:0db8:85a3. Setting every remaining bit to zero and compressing the zeros gives 2001:0db8:85a3::/48. The rightmost hextets belong to the subnet and interface parts, not to a /48 prefix.

  6. Frame 006 of 015

    Which type of IPv6 address is primarily used for one-to-one communication across the global internet?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      Unique local addresses stay inside an organization; they are not routed on the global internet.

    2. Note on B

      Correct: global unicast addresses (currently 2000::/3) are globally routable and identify one interface, the IPv6 version of a public IPv4 address.

    3. Note on C

      Link-local addresses are confined to a single link; routers never forward them, so they cannot cross the internet.

    4. Note on D

      Multicast is one-to-many delivery to a group (FF00::/8), not one-to-one.

    Best answer: B · why

    Global unicast addresses are IPv6 addresses that are globally routable and reachable on the IPv6 Internet. They are equivalent to public IPv4 addresses and are used for one-to-one communication across the global internet. They typically begin with the prefix 2000::/3.

  7. Frame 007 of 015

    What does the IPv6 route ‘::/0’ represent in a routing table, and why is it critical for end-to-end connectivity?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      Multicast uses FF00::/8 groups; ::/0 matches every address, not only multicast destinations.

    2. Note on B

      The IPv6 loopback is ::1/128, a single address; ::/0 is the opposite extreme and matches everything.

    3. Note on C

      Link-local addresses are FE80::/10 and are never routed; ::/0 is a catch-all for routable destinations.

    4. Note on D

      Correct: ::/0 has a zero-length prefix, so it matches any destination that lacks a more specific route and sends it to the default next hop.

    Best answer: D · why

    The ‘::/0’ is the IPv6 default route, which matches any IPv6 address that does not have a more specific route. It ensures that packets destined for unknown networks are forwarded to a default gateway.

  8. Frame 008 of 015

    A network engineer needs to configure several servers in different data centers to offer the same content efficiently to users worldwide. Which type of IPv6 address should they use to ensure that users connect to the nearest server?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      Multicast sends one stream to every member of a group; it does not pick the single nearest server.

    2. Note on B

      A global unicast address belongs to one interface, so every user would reach the same server wherever they are.

    3. Note on C

      Unique local addresses are not routed on the internet, so users worldwide could not reach them at all.

    4. Note on D

      Correct: anycast puts the same address on several servers, and routing delivers each user’s packets to the nearest one.

    Best answer: D · why

    With anycast, the same address is configured on servers in several locations and advertised from each site. Routing then delivers each user’s packets to the topologically nearest server, which is how many DNS and content services spread load. Global unicast reaches one interface, multicast reaches every member of a group, and unique local addresses are not routed on the internet.

  9. Frame 009 of 015

    How are IPv6 multicast addresses identified?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      2000::/3 is the global unicast range, used for internet-routable one-to-one addresses.

    2. Note on B

      Correct: every IPv6 multicast address begins with FF, which is the FF00::/8 block.

    3. Note on C

      FE80::/10 is link-local unicast, present on every IPv6 interface.

    4. Note on D

      FC00::/7 is the unique local unicast range, the IPv6 counterpart of private IPv4 space.

    Best answer: B · why

    IPv6 multicast addresses always start with the prefix FF00::/8. The first eight bits are set to 1, making it easy to identify multicast addresses in IPv6. These addresses are used to send packets to multiple destinations simultaneously.

  10. Frame 010 of 015

    Which command configures a static IPv6 address on an interface?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      This is an IPv4 address with the IPv4 command; it configures no IPv6 address at all.

    2. Note on B

      Correct: in interface configuration mode, ipv6 address followed by the address and prefix length, here 2020::10/64, assigns a static IPv6 address.

    3. Note on C

      ::/0 is the default-route prefix, not a host address, so it cannot be assigned to an interface.

    4. Note on D

      The ip address command is for IPv4 only; IOS rejects an IPv6 address after it.

    Best answer: B · why

    IOS configures a static IPv6 address in interface configuration mode with ipv6 address followed by the address and prefix length, for example ipv6 address 2020::10/64. The ip address command takes only IPv4 addresses, and ::/0 is the default-route prefix, which cannot be used as an interface address.

  11. Frame 011 of 015Choose 3

    Which three statements are true about IPv6? (Choose three.)

    Choose 3 options, then Check answer. Open the note under any option to see why.

    1. Note on A

      IPv6 is not inherently weaker than IPv4; it has its own risks, such as rogue router advertisements, but the protocol itself is not less secure.

    2. Note on B

      Correct: IPsec was designed into IPv6 from the start through its extension headers; current RFCs recommend support rather than require it.

    3. Note on C

      32-bit addresses belong to IPv4; IPv6 addresses are four times as long.

    4. Note on D

      Correct: an IPv6 address is 128 bits, written as eight 16-bit hextets.

    5. Note on E

      Correct: SLAAC lets a host build its own address from the prefix in router advertisements, with no DHCP server needed.

    Best answers: B, D and E · why

    IPv6 addresses are 128 bits long, written as eight hextets, against 32 bits for IPv4. Hosts can configure themselves with SLAAC, using the prefix in router advertisements. IPsec (AH and ESP) was designed into IPv6 from the start; RFC 8504 now recommends IPsec support rather than requiring it, so it is built in but not switched on automatically. IPv6 is not inherently less secure than IPv4.

  12. Frame 012 of 015Choose 2

    Which two are valid IPv6 addresses? (Choose two.)

    Choose 2 options, then Check answer. Open the note under any option to see why.

    1. Note on A

      10.0.0.0/8 is an IPv4 private network in CIDR notation, not an IPv6 address.

    2. Note on B

      This is invalid twice over: g is not a hexadecimal digit, and :: may appear only once in an address.

    3. Note on C

      Correct: eight hextets of four hex digits each, with nothing compressed, is the full form of a valid IPv6 address.

    4. Note on D

      Correct: this compressed form is valid, because :: appears once and stands in for the missing all-zero hextets.

    5. Note on E

      192.168.1.1 is a dotted-decimal IPv4 address.

    Best answers: C and D · why

    An IPv6 address has eight 16-bit hextets of hexadecimal digits (0 to 9, A to F) separated by colons. Leading zeros in a hextet may be dropped, and one run of all-zero hextets may be replaced by :: once per address. 2001:0db8:85a3:0000:0000:8a2e:0370:7334 is the full form and 2001:db8::ff00:42:8329 is a valid compressed one. 2001:db8::g1ff::1 uses a non-hex character and two ::, and the other two are IPv4.

  13. Frame 013 of 015

    A router interface with MAC address 001a.2b3c.4d5e is configured with ‘ipv6 address 2001:db8:1:1::/64 eui-64’. Which IPv6 address does the interface get?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      FFFE is in the right place, but the 7th bit was not flipped. The first byte 00 must become 02, so the hextet is 021a, written 21a.

    2. Note on B

      The inserted bytes are reversed. Modified EUI-64 inserts FF then FE after the third byte of the MAC, which gives 2bff:fe3c, not 2bfe:ff3c.

    3. Note on C

      FFFE is inserted in the wrong spot. It goes between the third and fourth bytes of the MAC, right after 2b, not after 3c.

    4. Note on D

      Correct. FFFE goes in the middle of the MAC, and flipping the 7th bit turns the first byte 00 into 02, giving the interface ID 021a:2bff:fe3c:4d5e.

    Best answer: D · why

    Modified EUI-64 (RFC 4291, Appendix A) builds a 64-bit interface ID from a 48-bit MAC. Split the MAC into its first and last three bytes, 00:1a:2b and 3c:4d:5e, and insert FF:FE between them: 00:1a:2b:ff:fe:3c:4d:5e. Then invert the 7th bit of the first byte, the universal/local bit: 00 becomes 02. The result is 021a:2bff:fe3c:4d5e, and with leading zeros dropped the address is 2001:db8:1:1:21a:2bff:fe3c:4d5e. Forgetting the flip and putting FFFE in the wrong place are the usual errors. Source: RFC 4291.

  14. Frame 014 of 015

    Which is a valid compressed form of the IPv6 address 2001:0db8:0000:0000:0000:00a0:0000:0001?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      Correct. The run of three zero hextets becomes ::, 00a0 loses only its leading zeros, and the lone zero hextet is written as 0.

    2. Note on B

      :: may appear only once in an address. With two, a reader cannot tell how many zero hextets each one replaces.

    3. Note on C

      Only leading zeros may be dropped. 00a0 becomes a0; dropping its trailing zero turns it into 000a, a different value.

    4. Note on D

      This writes only six hextets. Without :: to stand for the missing zero hextets, an address must show all eight.

    Best answer: A · why

    IPv6 text compression follows two rules. First, leading zeros in each 16-bit hextet may be dropped, but trailing zeros may not: 0db8 becomes db8 and 00a0 becomes a0, never a. Second, one run of consecutive all-zero hextets may be replaced by a double colon, and only once per address, or the address becomes ambiguous. RFC 5952 also recommends using :: for the longest zero run and lower-case hex. Applied here, the three zero hextets after db8 become ::, giving 2001:db8::a0:0:1. Sources: RFC 4291 and RFC 5952.

  15. Frame 015 of 015

    An engineer enters ‘ipv6 enable’ on a router interface but configures no global unicast address on it. Which statement about that interface is true?

    Pick an option. The verdict shows here; open the note under any option to see why.

    1. Note on A

      IPv6 requires a link-local address on every interface, and the router creates it as soon as IPv6 is enabled there.

    2. Note on B

      Correct. Every IPv6-enabled interface must have a link-local address, and ‘ipv6 enable’ creates one even with no global address.

    3. Note on C

      Unique local addresses must be configured or assigned. A router does not create one automatically when IPv6 is enabled.

    4. Note on D

      The unspecified address :: is used only in special cases, such as duplicate address detection, not as a normal source address.

    Best answer: B · why

    RFC 4291 requires every IPv6 interface to have at least one link-local unicast address, from fe80::/10. On Cisco IOS, ‘ipv6 enable’ turns IPv6 on for the interface and creates its link-local address automatically (usually with a modified EUI-64 interface ID), even if no global unicast address is configured. That address is enough for neighbor discovery and for routing protocols such as OSPFv3, which use link-local addresses as next hops. Unique local addresses (fc00::/7) must be configured, and the unspecified address :: is never a normal source address. Source: RFC 4291.

IPv6 in the v2.0 list, and the next stop

The two v1.1 topics here (1.8 for addressing and prefixes, 1.9 for address types) merge into one troubleshooting topic in v2.0 (1.4), and OSPFv3 brings dynamic IPv6 routing into topic 3.3, while IPv6 static routes sit on both lists; the v2.0 changeover page has the timing.

The next landmark is the switch port these addresses ride through: switching, VLANs and trunks. To repeat this area on its own, use the IPv6 filter on the practice test.

Primary sources

  1. [1]RFC 4291, IP Version 6 Addressing Architectureaddress types, interface IDs, modified EUI-64 (Appendix A). Checked October 29, 2026.
  2. [2]RFC 5952, A Recommendation for IPv6 Address Text Representationcompression and the preferred form. Checked October 29, 2026.
  3. [3]RFC 4193, Unique Local IPv6 Unicast AddressesChecked October 29, 2026.
  4. [4]RFC 4862, IPv6 Stateless Address AutoconfigurationChecked October 29, 2026.
  5. [5]Cisco Learning Network: CCNA v1.1 exam topicstopics 1.8 and 1.9. Checked October 29, 2026.
  6. [6]Cisco Learning Network: CCNA v2.0 exam topicstopic 1.4, from February 3, 2027. Checked October 29, 2026.

Related frames