MAC Address-Triggered ARP Entry Update
On an Ethernet network, a host sends and receives Ethernet data frames using MAC addresses. The Address Resolution Protocol (ARP) maps IP addresses to MAC addresses. When two devices on different network segments communicate with each other, they need to map IP addresses to MAC addresses and outbound interfaces according to ARP entries.
The outbound interfaces in matching MAC address and ARP entries are usually consistent. In Figure 2-7, the outbound interface in both the MAC address entry and ARP entry is GE1/0/1 at T1.
- Between T1 and T2, the interface for the entry changes.
- At T2, after a packet is received from a peer device, the outbound interface in the MAC address entry is changed to GE1/0/2. However, the outbound interface in the ARP entry remains GE1/0/1.
- At T3, the ARP entry expires, and the outbound interface in the ARP entry is changed to GE1/0/2 through an ARP aging probe. Between T2 and T3, GE1/0/1 is unavailable, meaning communication between devices on different network segments is interrupted.
MAC address-triggered ARP entry update enables a device to update the outbound interface in an ARP entry immediately after the outbound interface in the corresponding MAC address entry changes. In Figure 2-8, MAC address-triggered ARP entry update is enabled. At T2, after the outbound interface in the MAC address entry is changed to GE1/0/2, the outbound interface in the ARP entry is immediately changed to GE1/0/2. This prevents communication interruptions encountered in the previous example.
The MAC address-triggered ARP entry update function is often used on networks where devices in a Virtual Router Redundancy Protocol (VRRP) group connect to servers (for more information, see Configuring MAC Address-Triggered ARP Entry Update to Improve VRRP Switchover Performance), or Layer 3 traffic switching scenarios where STP and Smart Link are used.