A comprehensive reference guide to IPv6 terminology, acronyms, and technical concepts. This glossary provides clear, concise definitions for the protocols, technologies, address types, and mechanisms that form the foundation of IPv6 networking.
The process of discovering the link-layer (MAC) address of a neighbor on the same network link, given its IPv6 address. In IPv6, this function is performed by the Neighbor Discovery Protocol (NDP), replacing IPv4's Address Resolution Protocol (ARP).
A network addressing and routing method where packets sent to a single IPv6 address are delivered to the nearest device (according to routing protocol metrics) among a group of potential receivers that share that address. Anycast addresses use the same address space as global unicast addresses (2000::/3) and are commonly used for DNS services, CDNs, and high-availability services.
Example: All Internet root nameservers use anycast addressing to route queries to the geographically nearest server.
The IPv4 protocol used to map network layer addresses to link-layer addresses. In IPv6, ARP is replaced by the Neighbor Discovery Protocol (NDP), which uses ICMPv6 messages instead of broadcast transmissions. See ARP replacement in IPv6 for more details.
The standardized exterior gateway protocol used to exchange routing and reachability information between autonomous systems on the Internet. BGP supports both IPv4 and IPv6, with multiprotocol extensions (MP-BGP) enabling IPv6 prefix advertisement and routing.
A connectivity state where IPv6 is configured on a device but connections time out or fail, often due to misconfiguration, firewall issues, or ISP problems. This condition can cause slower website loading as browsers attempt IPv6 connections before falling back to IPv4. Testing tools like test-ipv6.run can detect this problematic state.
An IPv6 address with an interface identifier generated from a cryptographic hash of a public key and other parameters, as defined in RFC 3972. CGAs are used in SEcure Neighbor Discovery (SEND) to verify address ownership without requiring a public key infrastructure. The address is cryptographically bound to the node's public key, preventing address spoofing.
A method for allocating IP addresses and routing that replaces the older classful network design. In IPv6, CIDR notation uses a slash followed by the prefix length (e.g., 2001:db8::/32) to indicate the network portion of an address.
A mechanism defined in NDP (RFC 4862) that ensures an IPv6 address is unique on a network segment before it is assigned to an interface. During DAD, a node sends a Neighbor Solicitation message with the tentative address as the target. If no Neighbor Advertisement is received within a specified timeout (typically 1 second), the address is considered unique and can be used.
A network protocol that provides stateful address configuration and other network parameters to IPv6 hosts. See DHCPv6 Configuration Guide for practical setup instructions. DHCPv6 comes in two flavors:
DHCPv6 and SLAAC are complementary and can be used together.
A DNS server mechanism that synthesizes AAAA records from A records, enabling IPv6-only clients to access IPv4-only services when combined with NAT64. When an IPv6-only client queries for a hostname and only an A record exists, DNS64 creates a synthetic AAAA record by embedding the IPv4 address in a special IPv6 prefix (typically 64:ff9b::/96).
A suite of IETF specifications for securing DNS information. DNSSEC works with both IPv4 and IPv6, providing origin authentication and data integrity for DNS queries and responses through cryptographic signatures.
An IPv6 address block reserved exclusively for use in documentation, examples, and sample code. Similar to IPv4's 192.0.2.0/24, this prefix should never appear in production networks or be routed on the Internet. Defined in RFC 3849.
Example: 2001:db8:85a3::8a2e:370:7334
An IPv6 transition mechanism defined in RFC 6333 that enables IPv4 service continuity over IPv6-only networks. DS-Lite uses IPv4-in-IPv6 tunneling to transport IPv4 traffic through an IPv6 network to a centralized IPv4-IPv4 NAT gateway (AFTR - Address Family Transition Router).
A network configuration where devices and infrastructure support both IPv4 and IPv6 simultaneously. See Dual-Stack Networking Guide for detailed implementation guidance. Dual stack is considered the most straightforward transition strategy, allowing gradual migration to IPv6 while maintaining IPv4 compatibility. Applications can use either protocol depending on availability and preference.
A method of generating the 64-bit interface identifier portion of an IPv6 address from a 48-bit MAC address. See EUI-64 Generation Guide for detailed explanation. The process:
Example: MAC 00:11:22:33:44:55 becomes interface ID 0211:22ff:fe33:4455
While simple to implement, EUI-64 has privacy implications as it allows device tracking across networks. Privacy Extensions (RFC 4941) were developed to address this concern.
Optional headers that follow the main IPv6 header and provide additional functionality. Extension headers include:
Extension headers enable modular packet processing but have raised security concerns, leading some networks to block packets containing them.
A 20-bit field in the IPv6 header used to identify packets belonging to the same traffic flow, enabling Quality of Service (QoS) and traffic engineering. All packets with the same source, destination, and flow label should receive identical routing treatment. Initially underutilized, flow labels are increasingly important for IPv6 QoS implementations.
The process of breaking large packets into smaller pieces to fit the Maximum Transmission Unit (MTU) of a network path. Unlike IPv4 where routers can fragment packets, IPv6 fragmentation is performed only by the sending host. If a packet exceeds the path MTU, intermediate routers drop it and send an ICMPv6 "Packet Too Big" message back to the source, which then performs fragmentation if necessary.
An IPv6 address that is globally routable on the Internet, equivalent to IPv4 public addresses. Global unicast addresses:
Example: 2001:0db8:85a3:0000:0000:8a2e:0370:7334
When you test your connectivity at test-ipv6.run, the IPv6 address displayed is your global unicast address.
Rules for shortening IPv6 address notation to improve readability:
:::: notation can only be used once per address to avoid ambiguityExample: 2001:0db8:0000:0000:0000:0000:0000:0001 becomes 2001:db8::1
An 8-bit field in the IPv6 header that limits packet lifetime by specifying the maximum number of router hops allowed. Each router decrements this value by one; when it reaches zero, the packet is discarded and an ICMPv6 "Time Exceeded" message is sent back to the source. Functionally equivalent to IPv4's Time To Live (TTL) field.
A DHCPv6 construct used to assign one or more non-temporary IPv6 addresses to a client interface. An IA-NA contains the assigned addresses along with their preferred and valid lifetimes. Clients can request multiple IA-NAs if they need addresses on different interfaces.
A DHCPv6 construct used for prefix delegation, where a delegating router assigns one or more IPv6 prefixes to a requesting router. The requesting router can then subdivide and assign these prefixes to its downstream networks. Common in ISP environments where customer premises equipment (CPE) receives a /56 or /60 prefix for internal use.
The organization responsible for coordinating global IP address allocation, DNS root zone management, and protocol parameter assignment. IANA allocates large IPv6 address blocks to Regional Internet Registries (RIRs), which further distribute them to ISPs and organizations.
An integral part of IPv6 defined in RFC 4443 that provides error reporting, diagnostic functions, and core protocol functions. Unlike ICMPv4, ICMPv6 also encompasses functions previously handled by separate protocols in IPv4:
ICMPv6 is essential for IPv6 operation and should never be completely blocked by firewalls.
A node's physical or logical attachment point to a network link. In IPv6, addresses are assigned to interfaces rather than nodes. A single physical interface may have multiple IPv6 addresses of different types (link-local, global unicast, etc.).
The rightmost 64 bits of an IPv6 address that uniquely identifies an interface on a link. Interface identifiers can be generated through various methods:
A protocol suite for securing IP communications through authentication and encryption. While optional in IPv4, IPsec support was initially mandatory in IPv6 implementations (RFC 2460), though this requirement was relaxed in RFC 6434. IPsec provides:
An IPv6 transition mechanism defined in RFC 5214 that creates automatic tunnels for transmitting IPv6 packets over IPv4 infrastructure within a site. ISATAP embeds IPv4 addresses within IPv6 addresses using the format: [64-bit prefix]:0000:5efe:[IPv4 address]. Primarily used in enterprise environments during IPv6 transition.
An IPv6 packet with a payload larger than 65,535 bytes (the maximum size possible with the standard 16-bit Payload Length field). Jumbo payloads are enabled through the Hop-by-Hop Options extension header and can reach up to 4,294,967,295 bytes. Also called "jumbograms," these are defined in RFC 2675 and used primarily in high-performance networks with large MTUs.
A communication medium over which network nodes can communicate at the link layer (Layer 2). Examples include Ethernet segments, Wi-Fi networks, and point-to-point connections. IPv6's link-local addresses are valid only within a single link.
An IPv6 unicast address with prefix FE80::/10 that is valid only for communication on a single network link. See Link-Local Addresses Function for detailed information. Link-local addresses:
fe80::1%eth0Example: fe80::1ff:fe23:4567:890a
Link-local addresses enable local network communication even when global addresses are unavailable.
The IPv6 equivalent of IPv4's 127.0.0.1, used by a node to send packets to itself. The loopback address is always ::1 and enables local testing and inter-process communication without sending packets onto the network.
An IPv6 transition mechanism defined in RFC 7597 that uses IPv4-in-IPv6 encapsulation and stateless port-restricted IPv4 address sharing. MAP-E allows ISPs to provide IPv4 service over an IPv6-only infrastructure by assigning shared IPv4 addresses with port ranges to customer equipment. The customer edge (CE) device performs encapsulation/decapsulation and port mapping.
An IPv6 transition mechanism defined in RFC 7599 that uses stateless IPv4-IPv6 translation instead of encapsulation. Like MAP-E, MAP-T enables IPv4 service over IPv6 infrastructure using shared IPv4 addresses with port restrictions. The key difference is that MAP-T translates packet headers rather than encapsulating them, potentially offering better compatibility with certain applications.
The largest packet size that can be transmitted over a network link without fragmentation. IPv6 requires all links to support an MTU of at least 1280 bytes (compared to 68 bytes for IPv4). Path MTU Discovery (PMTUD) determines the smallest MTU along a network path, enabling efficient packet sizing without fragmentation.
A method of sending IPv6 packets to multiple destinations simultaneously. Packets sent to a multicast address (prefix FF00::/8) are delivered to all interfaces that have joined that multicast group. IPv6 uses multicast extensively, replacing IPv4's broadcast mechanism with more efficient targeted group communication.
Common multicast addresses:
FF02::1 - All nodes on the local linkFF02::2 - All routers on the local linkFF02::1:2 - All DHCPv6 servers and relay agentsAn IPv6 protocol (part of ICMPv6) that manages multicast group membership on a link. MLD is equivalent to IPv4's IGMP (Internet Group Management Protocol) and allows hosts to inform routers about their multicast group memberships. Current version is MLDv2 (RFC 3810).
An ICMPv6 message (Type 136) used in the Neighbor Discovery Protocol to respond to Neighbor Solicitations or announce link-layer address changes. Neighbor Advertisements contain the target IPv6 address and link-layer address, enabling address resolution and neighbor unreachability detection.
A mechanism that allows IPv6-only clients to communicate with IPv4-only servers through stateful translation between IPv6 and IPv4 packets. See NAT64 and DNS64 Explained for comprehensive details. Typically deployed with DNS64, NAT64 translates IPv6 packets with addresses in a special prefix (often 64:ff9b::/96) to IPv4 packets. Defined in RFC 6146, NAT64 is crucial for IPv6-only network deployments.
A method of modifying IP address information in packet headers while in transit. While common in IPv4 to conserve addresses, NAT is generally unnecessary in IPv6 due to the vast address space. However, some organizations deploy IPv6 NAT for perceived security benefits, often using NPTv6 (Network Prefix Translation) defined in RFC 6296. See NAT in IPv6 Explained for more details.
A fundamental IPv6 protocol (RFC 4861) that enables nodes on the same link to discover each other, determine link-layer addresses, find routers, and maintain reachability information. See Neighbor Discovery Protocol Guide for detailed explanation. NDP replaces several IPv4 protocols including ARP, ICMP Router Discovery, and ICMP Redirect. It operates using five ICMPv6 message types:
A node attached to the same network link. In IPv6, neighbors communicate using link-local addresses and the Neighbor Discovery Protocol.
A table maintained by each IPv6 node that stores the link-layer addresses of neighbors on the same link, along with reachability state information. Similar to the ARP cache in IPv4, but with additional state tracking for Neighbor Unreachability Detection.
An NDP mechanism that tracks the reachability status of neighbor nodes. NUD detects when a neighbor becomes unreachable (due to failure, disconnection, or address change) and updates the neighbor cache accordingly. This prevents sending packets to unreachable destinations and enables quick recovery when paths fail.
An 8-bit field in the IPv6 header (and in extension headers) that identifies the type of header immediately following. Similar to IPv4's Protocol field, Next Header uses the same values (e.g., 6 for TCP, 17 for UDP, 58 for ICMPv6) plus additional values for IPv6 extension headers (e.g., 44 for Fragment Header, 43 for Routing Header).
Any device implementing IPv6, including hosts and routers. A node may have multiple interfaces, each with its own IPv6 addresses.
A stateless, algorithmic translation mechanism for IPv6-to-IPv6 defined in RFC 6296. NPTv6 translates one IPv6 prefix to another, typically used at the edge of a network to translate internal Unique Local Addresses (ULAs) to provider-assigned global addresses. Unlike NAT44, NPTv6 is stateless and maintains end-to-end address transparency for the interface identifier portion.
An ICMPv6 message (Type 135) used in the Neighbor Discovery Protocol for three purposes:
Neighbor Solicitations are sent to solicited-node multicast addresses, making them more efficient than IPv4 ARP broadcasts.
The first 24 bits of a MAC address that identify the manufacturer. OUIs are visible in EUI-64 interface identifiers, which can enable device fingerprinting and tracking—one reason Privacy Extensions were developed.
An IPv6 header plus payload. May include extension headers between the main header and the payload.
The process of determining the smallest MTU along a network path. In IPv6, PMTUD is defined in RFC 8201 and works differently than in IPv4:
PMTUD prevents fragmentation by ensuring packets are sized appropriately for the entire path.
A 16-bit field in the IPv6 header specifying the length of the packet payload in bytes, including extension headers but excluding the main IPv6 header. Maximum value is 65,535 bytes; larger payloads require the Jumbo Payload option.
A mechanism defined in RFC 8415 that allows a delegating router to assign IPv6 prefixes to requesting routers via DHCPv6. The requesting router can then subdivide the delegated prefix for use on its downstream networks. Common in ISP deployments where CPE routers receive prefixes (typically /56 or /60) that they further allocate to internal subnets.
Example: ISP delegates 2001:db8:1234::/56 to home router, which creates /64 subnets for different VLANs.
See Path MTU Discovery.
The network portion of an IPv6 address, typically expressed in CIDR notation. Common prefix lengths include:
Example: In 2001:db8:1234:5678::/64, the prefix is 2001:db8:1234:5678 with length 64 bits.
A mechanism that generates temporary IPv6 addresses with randomly generated interface identifiers instead of using EUI-64. Privacy Extensions addresses:
Also called "temporary addresses," these provide privacy while maintaining stable addresses for services that need them. See IPv6 Address Changes Explained for more details.
Mechanisms for prioritizing network traffic to ensure performance for critical applications. IPv6 supports QoS through:
The Flow Label provides more sophisticated flow identification than possible in IPv4.
An ICMPv6 message (Type 134) sent periodically by routers or in response to Router Solicitations. Router Advertisements contain critical configuration information including:
RAs are essential for IPv6 autoconfiguration and network operation.
A security feature defined in RFC 6105 that protects against rogue Router Advertisement attacks. Implemented on switches, RA Guard:
Early RA Guard implementations could be bypassed using fragmentation; modern versions address this vulnerability.
An ICMPv6 message (Type 137) sent by routers to inform hosts of better routing paths. A router sends a Redirect when it receives a packet from a host and determines that either:
Redirects optimize routing by allowing hosts to update their routing decisions dynamically.
Organizations responsible for allocating and registering IP addresses within specific regions:
RIRs receive IPv6 allocations from IANA and distribute them to ISPs and organizations in their regions.
A node that forwards IPv6 packets not explicitly addressed to itself. Routers send Router Advertisements and respond to Router Solicitations, enabling network autoconfiguration.
An ICMPv6 message (Type 133) sent by hosts to request immediate Router Advertisements from routers. When a host connects to a network, it multicasts an RS to FF02::2 (all-routers) to expedite configuration rather than waiting for the next periodic RA.
An IPv6 extension header (Next Header = 43) that specifies a list of intermediate nodes to visit along the packet's path. Type 0 Routing Headers were deprecated due to security concerns (RFC 5095). Type 2 is used for Mobile IPv6 route optimization, while Segment Routing IPv6 (SRv6) uses Type 4.
Defines the reachability domain of an IPv6 address. Common scopes include:
Scope is particularly important for multicast addresses, where it's encoded in the address itself.
An identifier added to link-local and site-local addresses to specify which interface or zone the address belongs to. Required because link-local addresses are not unique across different links. Represented as a percent sign followed by the interface name or index.
Example: fe80::1%eth0 or fe80::1%3
An extension to NDP defined in RFC 3971 that provides cryptographic security for NDP messages through:
SEND protects against NDP spoofing and man-in-the-middle attacks but faces deployment challenges due to complexity and performance overhead.
A tunneling mechanism that encapsulates IPv6 packets within IPv4 packets for transmission across IPv4 infrastructure. Unlike automatic mechanisms like 6to4, 6in4 tunnels are manually configured between specific endpoints. Also called "proto-41" because it uses IP protocol number 41.
An IPv6 transition mechanism defined in RFC 5969 that enables rapid IPv6 deployment across IPv4 infrastructure. Similar to 6to4 but using ISP-specific prefixes instead of 2002::/16, 6rd provides:
Successfully deployed by several large ISPs for quick IPv6 rollout.
An automatic tunneling mechanism defined in RFC 3056 that allows isolated IPv6 sites to communicate over IPv4 infrastructure. 6to4:
The well-known 6to4 relay router anycast address is 192.88.99.1.
A mechanism defined in RFC 4862 that enables IPv6 hosts to automatically configure addresses without DHCP servers. See SLAAC Explained for detailed configuration information. The SLAAC process:
SLAAC can operate alone or alongside DHCPv6, with RAs indicating which to use.
A special multicast address automatically created for each unicast address on an interface. Formed by taking the prefix FF02::1:FF00:0/104 and appending the last 24 bits of the unicast address.
Example: For unicast address 2001:db8::1234:5678, the solicited-node address is FF02::1:FF34:5678
Used for efficient address resolution in NDP, ensuring only nodes with matching addresses process Neighbor Solicitations.
A network architecture that uses the IPv6 routing header (Type 4) to encode network programs as ordered lists of instructions (segments). SRv6 enables:
Growing in importance for data center, 5G, and service provider networks.
Stateful protocols maintain information about client sessions (e.g., DHCPv6 tracking address assignments). Stateless protocols do not maintain session information (e.g., SLAAC, NPTv6). Stateless mechanisms are generally simpler and more scalable but offer less control and visibility.
An IPv6 transition technology defined in RFC 4380 that tunnels IPv6 packets over UDP through IPv4 NAT devices. Teredo:
Teredo should be disabled when native IPv6 is available.
See Privacy Extensions (RFC 4941). Temporary addresses use randomly generated interface identifiers that change periodically to prevent tracking.
An 8-bit field in the IPv6 header used for Quality of Service, similar to IPv4's Type of Service (ToS) or Differentiated Services (DS) field. The Traffic Class field enables routers to identify and prioritize different classes of traffic (e.g., voice, video, bulk data).
Technologies that enable IPv6 deployment and interoperation with existing IPv4 infrastructure. Categories include:
In IPv4, an 8-bit field limiting packet lifetime. IPv6 replaces TTL with the Hop Limit field, which serves the same function but with a more accurate name.
IPv6 addresses for private network use, analogous to IPv4 private addresses (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16). ULAs:
Example: fd12:3456:789a:1::1
Generated with random 40-bit Global ID to ensure uniqueness (RFC 4193).
An IPv6 address identifying a single interface. Packets sent to a unicast address are delivered to exactly one recipient. Unicast types include:
The all-zeros address (0:0:0:0:0:0:0:0 or ::) indicates the absence of an address. Used:
The period during which an IPv6 address is valid for use. Addresses have two lifetimes:
When the preferred lifetime expires, the address becomes deprecated (still usable for existing connections but not preferred for new ones). When the valid lifetime expires, the address must not be used.
IPv6 addresses reserved for specific purposes and defined in RFCs:
::1 - Loopback:: - UnspecifiedFF02::1 - All nodes on linkFF02::2 - All routers on linkFF02::1:2 - All DHCPv6 servers/relaysSee Scope ID. The interface identifier or zone index appended to link-local or site-local addresses to specify which scope zone the address belongs to.
Example: ping6 fe80::1%eth0
Understanding these terms becomes practical when testing and troubleshooting IPv6 connectivity. Visit test-ipv6.run to:
The comprehensive tests at test-ipv6.run evaluate your network across multiple connectivity scenarios, helping identify issues related to NDP, SLAAC, router advertisements, and other fundamental IPv6 protocols covered in this glossary.
For deeper exploration of specific topics, refer to these key RFCs:
IPv6 introduces a rich vocabulary of protocols, mechanisms, and addressing concepts. This glossary provides a foundation for understanding IPv6 networking, from basic address types to advanced transition mechanisms. As IPv6 deployment continues to grow globally, familiarity with these terms becomes increasingly valuable for network administrators, developers, security professionals, and anyone working with modern Internet infrastructure.
Whether you're configuring your first IPv6 network, troubleshooting connectivity issues, or simply exploring how the Internet's next-generation protocol works, these definitions provide a reference for the building blocks of IPv6 networking.
Last updated: 2025 - For IPv6 connectivity testing and more information, visit test-ipv6.run