ARP, the Address Resolution Protocol, answers a deceptively simple question on a local network: which hardware address belongs to this IPv4 address? When you ask what addresses are mapped by ARP, the short answer is that ARP maps IPv4 addresses to MAC addresses inside the same local network segment. That mapping lets a device deliver an IP packet to the correct network interface card without guessing. The process is automatic, fast, and mostly invisible, but it is fundamental to how Ethernet and IPv4 networks move traffic.
What ARP Maps: IPv4 to MAC
ARP creates a binding between two different kinds of addresses:
- Logical address: an IPv4 address, such as
192.168.1.25. - Physical address: a MAC address, such as
00:1A:2B:3C:4D:5E.
The IPv4 address identifies a host at the network layer. In real terms, the MAC address identifies a network interface at the data-link layer. ARP’s job is to connect those two identifiers so frames can be built and sent across a local link.
This means ARP does not map domain names to IP addresses. That is DNS. That's why aRP also does not map IPv6 addresses to MAC addresses in the same way; IPv6 uses the Neighbor Discovery Protocol instead. ARP is specifically an IPv4-era protocol, although it still appears in most IPv4 networks today.
Where the Mapping Is Stored: The ARP Cache
ARP does not ask the network for a MAC address every time it sends a packet. Instead, it stores recent mappings in a local ARP cache, sometimes called the ARP table. Each entry typically includes:
- The IPv4 address being resolved.
- The corresponding MAC address.
- The interface used to reach that address.
- The entry type, such as dynamic or static.
- An age or expiration timer.
Dynamic entries are learned automatically and expire after a period of inactivity. So static entries are added manually and remain until removed or the device restarts. Static ARP entries can be useful in small, stable networks, but they require maintenance and can become outdated if hardware changes.
How ARP Builds the Mapping
The ARP process follows a clear request-and-reply pattern.
- A host needs to send an IPv4 packet to another device on the same subnet.
- It checks its ARP cache for an existing mapping.
- If no mapping exists, it sends an ARP request as a broadcast frame: “Who has this IPv4 address? Tell me.”
- Every device on the local segment receives the broadcast, but only the device that owns the target IPv4 address responds.
- That device sends an ARP reply directly to the requester, including its MAC address.
- The requester stores the mapping in its ARP cache and uses it to build the Ethernet frame.
The request is broadcast because the sender does not yet know which MAC address should receive the question. The reply is usually unicast because the target now knows the requester’s MAC address from the request It's one of those things that adds up..
ARP Maps Local Next-Hop Addresses, Not Every Remote Address
A common misunderstanding is that ARP maps the final destination IP address for every packet. In practice, ARP resolves the next-hop address on the local link That's the part that actually makes a difference..
If Host A wants to reach Host B on the same subnet, ARP maps Host B’s IPv4 address to Host B’s MAC address. If Host A wants to reach a server on the internet, Host A does not ARP for the server’s public IP address. Instead, it ARPs for the default gateway’s IPv4 address and sends the frame to the gateway’s MAC address. The gateway then forwards the packet toward the destination.
So the addresses mapped by ARP are usually:
- Local host IPv4 addresses on the same broadcast domain.
- Router or default gateway IPv4 addresses on that local segment.
- Other local next-hop devices, such as load balancers or security appliances.
ARP does not resolve addresses across routers. Broadcast ARP requests stop at the local subnet boundary.
Special Cases: Proxy ARP and Gratuitous ARP
Two variations are worth knowing.
Proxy ARP allows a router or another device to answer ARP requests on behalf of a host that is not on the same link. The router replies with its own MAC address, and traffic is forwarded through it. Proxy ARP can help in some network designs, but it can also hide topology problems and create security risks if misused.
Gratuitous ARP is an unsolicited ARP message that a device sends to announce or update its own mapping. It is often used when an IP address moves to a new MAC address, when a failover occurs, or when a device wants to detect duplicate IP addresses. Gratuitous ARP helps neighboring devices update their ARP caches quickly Nothing fancy..
What ARP Does Not Map
To avoid confusion, keep these boundaries in mind:
- ARP does not map domain names. DNS handles name-to-IP resolution.
- ARP does not map IPv6 addresses. IPv6 uses Neighbor Discovery.
- ARP does not map remote destination IPs across routers. It resolves local next hops.
- ARP does not guarantee identity. A malicious device can send false ARP replies, a technique called ARP spoofing or ARP poisoning.
Because ARP has no built-in authentication, attackers on the same local network can claim that a certain IPv4 address belongs to their MAC address. This can redirect traffic through an attacker’s device. Network teams often use dynamic ARP inspection, port security, static ARP entries, and segmentation to reduce that risk That's the whole idea..
Why the Mapping Matters
Every IPv4 packet must eventually be