Troubleshooting Topology Errors

• Given a troubleshooting scenario involving a network with a particular physical topology (i.e. bus, star/hierarchical, mesh, ring, and wireless) and including a network diagram, identify the network area affected and the cause of the problem.

As discussed in Chapter 1, “Introduction to Networking,” several different topologies are used for networks. Each of these different types of network designs has failure points that are specific to the topology being used. To get a better idea of what is involved in troubleshooting these topology errors, the following sections provide specific troubleshooting scenarios and identify the potential causes of the problems.

Bus Network Errors

Recall from Chapter 1 that a bus topology connects all computer stations in a linear fashion. In the early days of Ethernet, the bus topology was the most widely used topology, and network administrators during that time were very experienced with the techniques involved in troubleshooting a bus network.

Before looking at specific bus troubleshooting scenarios, let's review the following characteristics of the bus topology:

• The cable used on a bus network has two distinct physical endpoints. Each of these cable ends requires a terminator. Terminators are used to absorb electronic signals so that they are not reflected back on the media, compromising data integrity. A failed or missing terminator will render the entire network segment unusable.

• The addition, removal, or failure of a device on the network might prevent the entire network from functioning. Also, the coaxial cable used in a bus network can be damaged very easily. Moving cables in order to add or remove devices can cause cable problems.

• A bus topology must be continuous. A break in the cable at any point will render the entire segment unusable. If the location of the break in the cable is not apparent, you can check each length of cable systematically from one end to the other to identify the location of the break, or you can use a tool such as a time domain reflectometer.

• One end of the bus network should be grounded. Intermittent problems or a high occurrence of errors may indicate poor or insufficient grounding.

Now that we have looked at some of the considerations for bus topologies, we can examine some possible troubleshooting scenarios and solutions.

Star Network Errors

A star network is the most commonly used network topology today. With a star topology, each computer connects to a centralized device, and each system requires its own cable. Troubleshooting a star network has different considerations from troubleshooting a bus topology. As a network administrator today, you can expect to troubleshoot star networks.

The following list contains a few of the main characteristics of the star topology and some pointers to potential steps you can take when troubleshooting a star network:

• Each device on the network requires its own cable, which is connected to a centralized device such as a hub. A cable failure should affect only the device connected to that cable length.

• Devices can be added or removed from a star network without affecting the existing users on the network. If other stations are affected by the addition or removal of devices, there might be a problem with a hub or switch.

• A centralized device provides a single point of failure in a star network. If a hub were to fail, for instance, all devices connected to it would be unable to access the network. When you know this and the fact that a cable problem should affect only a single system, you can significantly reduce the amount of time needed to isolate a problem in a star network.

• Hubs and switches have indicator lights, or LEDs, that show the states of connected devices as well as representations of network utilization and collision statistics. You can use the lights as a resource when troubleshooting a problem. No lights means no power. Most modern hubs and switches cannot operate without power.

• Hubs and switches can be connected to each other to provide more capacity on the network. If all the devices connected to one hub can see each other but not the rest of the network, you should suspect a problem with the hub-to-hub or switch-to-switch connection.

Ring Network Errors

A ring topology is not commonly used in today's network environments, but just in case you are working with a ring network, it is worthwhile knowing what to look for when troubleshooting one. The following is a review of the characteristics of a ring topology:

• The failure of a single computer or section of cable can cause complete network disruption.

• Only one computer can send data onto the network at a time, and the sending computer must have access to the token to send the data.

• Ring topologies are seen in Token Ring and Fiber Distributed Data Interface (FDDI) networks. Each of these systems uses fault-tolerant strategies.

Mesh Network Errors

A mesh topology offers high redundancy by providing several paths for data to reach its destination. In a true mesh network, each device on the network is connected to every other device, and if one cable fails, there is another to provide an alternative data path. Given the number of cables involved, a mesh network can be somewhat tricky to troubleshoot.

Let's review the characteristics of a mesh topology and look at some of the factors that can affect troubleshooting in mesh systems:

• A mesh topology interconnects all devices on the network, offering the highest level of redundancy of all the topologies. In a pure mesh environment, all devices are directly connected to all other devices. In a hybrid mesh environment, some devices are connected only to certain others in the topology.

• Although a mesh topology can accommodate failed links, mechanisms should still be in place so that failed links are detected and reported.

• Design and implementation of a true mesh network can be complex and often requires specialized hardware devices.

Mesh networks are so rare that it's unlikely that you will be faced with troubleshooting one but there will likely be questions on the Network+ exam that focus on mesh networks.

Wireless Network Errors

Wireless networks do not require physical cable to connect computers; rather, they use wireless media. The benefits of such a configuration are clear—users have remote access to files and resources without the need for physical connections. Wireless networking eliminates cable faults and cable breaks. It does, however, introduce its own considerations: signal interference and security.

The following list summarizes the characteristics of a wireless network:

• A wireless network allows for a remote connection without requiring cumbersome cabling.

• Users can be added to an existing network without disruption to current users.

• Common media types for wireless networks include infrared, radio waves, and satellite communication. The most common method used for wireless local area network (LAN) implementations is radio waves.

• It is possible to eavesdrop on wireless signals. Therefore, security must be carefully considered.

• Wireless communication has limited speed compared to cabled Ethernet networks.

• Some types of wireless communications require a point-to-point direct line-of-sight connection.