Packet sniffing is a technique of monitoring network traffic. It is effective on both switched and nonswitched networks. In a non-switched network environment packet sniffing is an easy thing to do. This is because network traffic is sent to a hub which broadcasts it to everyone. Switched networks are completely different in the way they operate.
Switches work by sending traffic to the destination host only. This happens because switches have CAM tables. These tables store information like MAC addresses, switch ports, and VLAN information. Before sending traffic from one host to another on the same local area network, the host ARP cache is first checked. The ARP cache is a table that stores both Layer 2 (MAC) addresses and Layer 3 (IP) addresses of hosts on the local network. If the destination host isn’t in the ARP cache, the source host sends a broadcast ARP request looking for the host. When the host replies, the traffic can be sent to it. The traffic goes from the source host to the switch, and then directly to the destination host. This description shows that traffic isn’t broadcast out to every host, but only to the destination host, therefore it’s harder to sniff traffic.
This paper discusses several methods that result in packet sniffing on Layer 2 switched networks. Each of the sniffing methods will be explained in detail. The purpose of the paper is to show how sniffing can be accomplished on switched networks, and to understand how it can be prevented.
Download the paper in PDF format here.