IPv6
Course Description
IPv6 is the successor of IPv4, and the main reason we need it is that we ran out of IPv4 addresses. IPv4 uses 32-bit addresses, which gives us about 4.3 billion addresses. IPv6 uses 128-bit addresses, so the address space is huge. But IPv6 is more than just longer addresses: there is no broadcast anymore, ARP is replaced by Neighbor Discovery, and hosts can configure their own address with SLAAC. In this course, you’ll learn how IPv6 addressing works and how to configure IPv6 on Cisco IOS routers and switches, including routing protocols, security features, tunneling, and NAT64.
Course Highlights
In this course you will learn:
- The differences between IPv4 and IPv6, and how to write and shorten IPv6 addresses.
- The IPv6 address types: global unicast, unique local, link-local, multicast, and anycast.
- How to find the prefix of an IPv6 address, create subnets, and summarize IPv6 prefixes.
- How Neighbor Discovery, EUI-64, SLAAC, and DHCPv6 work.
- How to configure IPv6 static routes, RIPng, EIGRP, OSPFv3, and MP-BGP.
- How to secure IPv6 with access-lists and First Hop Security features like RA Guard and DHCPv6 Guard.
- How to run IPv6 over an IPv4 network with manual, 6to4, 6RD, and ISATAP tunnels, and over MPLS with 6PE and 6VPE.
- How NAT64 and NPTv6 translate addresses and prefixes.
Presented to you by instructor Rene Molenaar, CCIE #41726
Course at a glance
- 49 lessons in 7 units.
- Format: self-paced lessons with Cisco IOS configurations, packet captures, and video.
- Last updated: October 2026.
Requirements
- No IPv6 knowledge required; we start from the beginning.
- You should be familiar with the IPv4 version of each topic: addressing, subnetting, routing protocols, and access-lists.
Course Schedule
What is IPv6?
IPv6 (Internet Protocol version 6) is the successor of IPv4. An IPv6 address is 128 bits long and written as eight groups of four hexadecimal characters, separated by colons, which we can shorten by removing leading zeros and replacing one string of zeros with a double colon. The first 64 bits are usually the prefix, and the last 64 bits identify the interface. The IPv6 header has a fixed size of 40 bytes, and only the source host fragments packets, not routers.
IPv6 isn’t just IPv4 with longer addresses. There is no broadcast: IPv6 uses multicast and a new address type, anycast. Every interface has a link-local address in the FE80::/10 range, and routing protocols use it to talk to their neighbors. ARP is replaced by NDP (Neighbor Discovery Protocol), which uses ICMPv6 and solicited-node multicast addresses. Hosts can configure their own address with SLAAC, using the prefix in a router advertisement, or get one from a DHCPv6 server.
Who uses IPv6? Mobile carriers, ISPs, cloud providers, and large content networks have deployed it widely, and enterprise networks are following. Since IPv4 and IPv6 will run side by side for years, network engineers need both: dual stack, tunneling, and translation techniques like NAT64.
IPv6 Learning Roadmap
This roadmap follows the course units. Each topic has a short explanation and links to the lessons where we cover it in detail.
IPv6 Addressing
We start by comparing IPv4 and IPv6, and learn how to write IPv6 addresses and how to shorten them. There are different address types: global unicast addresses that you can use on the Internet, unique local addresses that are similar to private IPv4 addresses, link-local addresses, multicast, and anycast. You’ll also learn how to find the prefix of an address, how prefixes are assigned from IANA down to your subnets, how to summarize them, and how EUI-64 creates an interface ID from a MAC address.
Neighbor Discovery and Address Autoconfiguration
NDP replaces ARP. To find the MAC address of a neighbor, a host sends a neighbor solicitation to the solicited-node multicast address of that neighbor, so only the right host has to process it. Routers send router advertisements with the prefix, which hosts use for SLAAC, and the router advertisement preference tells hosts which router to prefer when there is more than one. A DHCPv6 server can assign addresses with stateful DHCPv6, or only extra information like DNS servers with stateless DHCPv6, and prefix delegation hands out whole prefixes to customer routers. A general prefix lets you define a prefix once and reuse it on all your interfaces, and when a host doesn’t get an address, you’ll learn how to troubleshoot SLAAC.
IPv6 Routing
Before a Cisco router routes IPv6 packets, you need the ipv6 unicast-routing command. We start with static routes and then configure the IPv6 versions of the routing protocols you already know: RIPng, EIGRP for IPv6, and OSPFv3. OSPFv3 works much like OSPFv2, with a few differences: it runs per link, uses link-local addresses, has new LSA types, and is enabled on the interface instead of with network commands. You’ll also learn how OSPFv3 advertises a default route. For BGP, we use MP-BGP to advertise IPv6 prefixes and route filtering to control which ones we accept and advertise.
Troubleshooting IPv6 Routing and Redistribution
When IPv6 routing doesn’t work, the problem is often the same as with IPv4: neighbors that don’t form an adjacency, or routes that aren’t advertised. We troubleshoot RIPng and OSPFv3 neighbor adjacencies, configure redistribution between RIPng and OSPFv3, and fix redistribution problems. We also look at OSPFv3 prefix suppression to keep transit link prefixes out of the routing table.
IPv6 Security
IPv6 access-lists are always named and extended, and they end with invisible statements that permit neighbor discovery before the final deny. OSPFv3 has no authentication field of its own and uses IPsec for authentication and encryption. On switches, the First Hop Security features protect hosts: RA Guard blocks rogue router advertisements, DHCPv6 Guard blocks rogue DHCPv6 servers, and device tracking builds the binding table that ND inspection and Source Guard depend on. A port ACL applies an IPv6 access-list to a switch port.
IPv6 Tunneling
When there is an IPv4 network between your IPv6 sites, you can tunnel IPv6 through it. Manual and GRE tunnels are point-to-point and support IGPs, and you can protect a GRE tunnel with IPsec. Automatic 6to4 tunnels and ISATAP embed the IPv4 address of the tunnel endpoint in the IPv6 address, and 6RD improves 6to4 by letting ISPs use their own prefix. When a 6to4 tunnel doesn’t come up, you’ll learn how to troubleshoot it. Service providers with an IPv4 MPLS core can offer IPv6 with 6PE and 6VPE.
NAT64, NPTv6, Multicast and Services
NAT64 lets IPv4-only and IPv6-only hosts talk to each other by translating both addresses and protocols. NPTv6 is different: it only translates the prefix, one to one, without overloading. We finish with IPv6 multicast using BSR and an RP, and HSRP for IPv6, which requires HSRP version 2 and uses a virtual link-local address as the default gateway.
Frequently Asked Questions
Yes. Mobile carriers, many ISPs, cloud providers, and large content networks run IPv6 at scale, so a big part of Internet traffic already uses it. Enterprise networks are slower to adopt it, because NAT keeps IPv4 working and migrating takes effort. That’s why you’ll see IPv4 and IPv6 side by side for years. Compare IPv4 and IPv6.
Eventually, but not anytime soon, and there’s no date when IPv4 will be switched off. For now, most networks run dual stack, and translation techniques like NAT64 let IPv6-only and IPv4-only hosts talk to each other. As a network engineer, you need to know both. Learn how NAT64 works.
In most cases, leave it on. Modern operating systems prefer IPv6 when it’s available, and disabling it can break some features. If your ISP doesn’t offer IPv6, everything simply works over IPv4. Just keep in mind that IPv6 traffic needs its own firewall rules, since there’s no NAT to hide behind. Learn about the IPv6 address types.
Normally not. There are so many IPv6 addresses that every device can have a globally unique one, so you don’t need NAT to save addresses. You use a firewall to control what can reach your hosts instead. There’s NPTv6 for prefix translation and NAT64 for talking to IPv4-only hosts, but they solve different problems. Learn how NPTv6 works.
Yes. The CCNA 200-301 covers IPv6 addressing and prefixes, address types, shortening, EUI-64, and IPv6 static routes, so don’t skip it. If you understand IPv4 subnetting and can count in hexadecimal, IPv6 is much easier than it looks. Read our introduction to IPv6.
IPv6 is the successor of IPv4. It uses 128-bit addresses instead of 32-bit addresses, so there are enough addresses for every device. IPv6 also has no broadcast, replaces ARP with Neighbor Discovery, and lets hosts configure their own address with SLAAC. Read our introduction to IPv6.
IPv4 uses 32-bit addresses in dotted decimal notation, and IPv6 uses 128-bit addresses in hexadecimal. The IPv6 header is a fixed 40 bytes and has no checksum, and only the source host fragments packets. IPv6 has no broadcast, uses NDP instead of ARP, and supports SLAAC next to DHCPv6. Compare IPv4 and IPv6.
IPv6 has unicast, multicast, and anycast addresses, but no broadcast. The main unicast types are global unicast addresses for the Internet, unique local addresses (FC00::/7) for private use, and link-local addresses (FE80::/10) that only work on the local link. Multicast addresses use the FF00::/8 prefix. Learn about the IPv6 address types.