Problem 7 - An Introduction to Computer Networks

Question

Suppose two Ethernet switches are connected in a loop as follows; S1 and S2 have their interfaces 1 and 2 labeled. These switches do not use the spanning- tree algorithm.Suppose A attempts to send a packet to destination B, which is unknown. S1 will therefore flood the packet out interfaces 1 and 2. What happens then? How long will A's packet circulate?

Step-by-Step Solution

Answer

In the event S1 sends a packet to an unknown destination B, the packet is flooded out of both interfaces, leading to an endless circulation of the packet between S1 and S2. This will continue indefinitely, leading to a packet storm scenario and potentially disrupting network operations due to bandwidth overconsumption.

Step 1: Understand the network architecture

Here, we have two Ethernet switches, S1 and S2, connected in a loop. Network node A tries to send a packet to another node B, but B's location is unknown. As a result, switch S1 will flood the packet out of both of its interfaces (1 and 2).

Step 2: Track the path of the packet

After S1 floods the packet, either one of the links leading from S1 may lead back to S1 through S2. Since Ethernet switches forward packets to all ports (except for the receiving port), the packet will keep circling indefinitely between S1 and S2.

Step 3: Describe the problem that arises

This creates a packet storm in the ethernet loop, as there is no mechanism in place to 'age out' or remove the circulating packet. The packet will keep circulating between the Ethernet switches indefinitely, consuming bandwidth and potentially disrupting the network until manually removed or the switches are restarted.

Key Concepts

Network Loops

Imagine you are rolling a ball around a looped track. Similarly, in computer networks, a network loop occurs when packets travel in an endless circle between devices, such as switches in a network.
This usually happens due to improper configurations, causing packets to circulate without reaching their destination or being terminated.
Without intervention, this loop can cause network congestion and lead to various performance issues. It is essential to prevent network loops to ensure smooth and efficient communication in a network.

Packet Flooding

Packet flooding in a network refers to when a switch sends out a packet to all of its interfaces, except the one it received the packet on. This process is known as broadcasting, and is used to find the destination device when its location is unknown.
While useful, packet flooding can lead to excessive network traffic if not managed properly.

• It can burden network resources, as it may result in multiple copies of the same packet circulating through the network.
• This becomes a serious issue in the presence of network loops, where flooded packets can circle indefinitely, causing network storms.

Effective network management must strike a balance to prevent packet flooding from overwhelming the network.

Spanning Tree Algorithm

The Spanning Tree Algorithm (STA) is crucial in preventing network loops.
This algorithm works by organizing network connections into a tree structure, where loops are identified and disabled, while keeping one path open for data to flow.
Here's how it works:

• The STA identifies all possible paths within the network and selects the best one to use, effectively avoiding loops.
• If a primary path fails, the STA reconfigures the network to activate a backup path.

This approach helps maintain network stability and ensures data packets reach their destination efficiently.
In networks using the Spanning Tree Algorithm, the risk of packet flooding leading to network congestion due to loops is significantly reduced.