4.7. Troubleshooting Various Network Topologies

Depending on the type of network topology that you are using, you might experience a network failure that is specific to that topology. It's important that you understand the basics of each topology so that you can troubleshoot any type of network topology. In this section, we discuss each of the four main network topologies: bus, star, mesh, and ring.

4.7.1. Critical Information

Given a troubleshooting scenario involving a network with a particular network topology, you should be able to identify the network area affected and the cause of the stated failure. You should be familiar with these concepts for the four main network topologies: bus, star, ring, and mesh.

4.7.1.1. Bus

As we discussed in Chapter 1, the bus topology using coaxial cable became very popular in networks in the late 1980s and early 1990s, before the more sophisticated star topology began to emerge. It used an Ethernet baseband technology and was often referred to as 10BASE 2 or thinnet when thin coaxial cable was used, and as 10BASE 5 or thicknet when thicker coaxial cable was used. The main advantage of the bus topology was that relatively small amounts of coaxial cable could be used to connect computers. The bus topology uses T connectors to connect a signal to a computer and let it run through the T connection at the same time. These T connectors enable the network to continue to function even if one of the computers in the bus should fail or shut down.

The main disadvantage of bus networks was that they were not very fast, only 10Mbps, and they were not very reliable. The chief reason that bus networks are not reliable is that they require a terminator (or resistor) at each end of a segment, which absorbs signal and prevents signal bounce. This means that if the coaxial cable develops a break anywhere in a segment, then essentially two networks are created, neither of which has the required terminators on each end; therefore, all communication in the entire segment will fail. For these reasons, the bus topology has been replaced by the star topology in most networks.

4.7.1.2. Star

As we discussed in Chapter 1, the star topology is the most common network topology in use today. With the star topology, each computer has its own cable, which is connected to a central hub or switch. This means that a failure in a single computer or even in a single computer's cable will not affect the rest of the network segment. A failure in the hub or switch, however, will affect all of the computers that are connected to it. For this reason, most companies use multiple hubs and switches in a network design to provide alternate paths for client server communication. Also, many companies keep replacement hubs and/or switches standing by in case the primary one should fail.

4.7.1.3. Mesh

A mesh topology is probably the easiest topology to troubleshoot, because a failure of one connection still leaves other connections available to be used for troubleshooting the failure. As we discussed in Chapter 1, a full mesh provides a connection from every component in the mesh to every other component in the mesh. This creates a tremendous degree of fault tolerance.

Typically a mesh topology is not used to connect individual computers, but instead to connect networks with redundant connections. If two servers in a full mesh cannot communicate, the failure is more than likely with the one of the servers rather than with the connection, since there are multiple connections on which to communicate.

4.7.1.4. Ring

Ring topologies are not used very often today, because they have been replaced by more reliable and faster star topologies with Ethernet switches. Though they are not common, you still might need to troubleshoot a ring topology network and you might very well be asked about one on the test.

As we discussed in Chapter 1, computers are logically arranged in a circle (or ring) and a spe-cial signal called a token is circulated around the ring. When a computer wants to communicate on the ring, it must take possession of the token and attach its message. In IBM Token Ring networks, the token must stay attached the message until the message is delivered and acknowledged. In FDDI networks, the message can be sent and the token can move down the ring and be used by another computer.

In an IBM token ring network a break in the ring will result in all computers not being able to communicate. On the other hand, FDDI networks offer fault tolerance in the form of a dualring design consisting of a primary ring and a secondary ring. Ring topology networks also have special packets that detect a slow or failing network. In some ring networks a process called beaconing is automatically used by the system to determine where the break occurred.

4.7.2. Exam Essentials

Describe troubleshooting basics of the bus topology. The bus topology is a coaxial cable–based network topology that is not used very often today. The main disadvantage of the bus topology is that, since it requires terminator at each end of a segment, a break anywhere in network segment will cause all communication among all computers in the segment to fail. Because of the use of T connectors, a failed or shut-down computer in a segment will not affect communication of the other computers in the segment.

Know troubleshooting basics for the star topology. The star topology provides a separate network cable for each computer in the network. All of the computers are connected to a central hub or switch. A failure of a computer or even a break in a single network cable will not have an effect on all of the other computers in the network segment. However, a failure of the central hub or switch will have an affect on communication of all computers in the network segment.

List troubleshooting basics of the mesh topology. A mesh topology provides the greatest amount of fault tolerance of any network topology. A mesh topology is typically used between networks and not between individual computers. A fullmesh topology provides a separate connection for every component in a mesh to every other component in the mesh. If a server in a full mesh cannot communicate with another server in the same full mesh, the fault most likely lies with the servers and not with the connections.

Describe troubleshooting basics of the ring topology. A ring topology consists of computers that are logically arranged in a ring. A special packet called a token is circulated around the ring and the computers must have the token in order to communicate on the network. There are two main types of rings: IBM Token Ring and FDDI. IBM Token Rings typically do not offer fault tolerance, so a break in the ring will stop all communication. In contrast, FDDI networks are built on a dualring topology with a secondary ring that is used to provide fault tolerance should the primary ring fail.