Troubleshooting Wiring- and Infrastructure-Related Problems

• Given a network troubleshooting scenario involving a wiring/infrastructure problem, identify the cause of the problem (e.g. bad media, interference, network hardware).

You will no doubt find yourself troubleshooting wiring and infrastructure problems considerably more infrequently than you'll troubleshoot client connectivity problems—and thankfully so. Wiring-and infrastructure-related problems can be very difficult to trace, and sometimes a very costly solution is needed to remedy the situation. When troubleshooting these problems, a methodical approach is likely to pay off.

Wiring problems are related to the actual cable used in a network. For the purposes of the Network+ exam, infrastructure problems are classified as those related to network devices such as hubs, switches, and routers.

Troubleshooting Wiring

Troubleshooting wiring involves knowing what wiring your network uses and where it is being used.

Determining Your Wiring

As mentioned in Chapter 2, “Cabling and Connectors,” the cable used has certain limitations, in terms of both speed and distance. It might be that the network problems are the result of trying to use a cable in an environment or a way for which it was not designed. For example, you might find that a network is connecting two workstations that are 130 meters apart with Category 5 UTP cabling. Category 5 UTP is specified for distances up to 100 meters, so exceeding the maximum cable length could be a potential cause of the problem.

Determining the type of cable used by a network is often as easy as reading the cable. The cable should be stamped with its type—whether it is, for example, UTP Category 5, RG-58, or something else. As you work with the various cable types used to create networks, you'll get to the point where you can easily identify them. However, you should be careful when identifying cable types because some cable types are almost indistinguishable. After you have determined the cable being used, you can compare the characteristics and limitations of that cable against how it is being used on the network.

Where the Cable Is Used

Imagine that you have been called in to track down a problem with a network. After some time, you discover that clients are connected to the network via standard UTP cable run down an elevator shaft. Recall from Chapter 2 that UTP has very poor resistance to electromagnetic interference (EMI) and therefore UTP and the electrical equipment associated with elevators react to each other like oil and water. The same can be said of cables that are run close to fluorescent light fittings. Such problems might seem farfetched, but you would be surprised at just how many environments you will work in that have random or erratic problems that users have lived with for a long time and not done anything about.

Part of troubleshooting wiring problems is to identify where the cable is run, to isolate whether the problem is a result of crosstalk or EMI. You need to be aware of problems associated with interference and the distance limitations of the cable being used.

If you find a problem with a network's cable, there are various things you can do to correct the problem. For cables that exceed the maximum distance, you can use a repeater to regenerate the signal, try to reroute the cable over a more economical route, or even replace the type of cable with one that has greater resistance to attenuation. The method you choose will often depend on the network's design and your budget.

For cable that is affected by EMI or other interference, you should consider replacing the cable with cable that is more resistant to such interference or rerouting the cable away from the source of the interference. If you do reroute cable, pay attention to the maximum distance, and make sure that as you're curing one problem you don't create another.

Troubleshooting the Infrastructure

If you are looking for a challenge, troubleshooting infrastructure problems is for you. It is often not an easy task, and it usually involves many processes, including baselining and performance monitoring. Only by using a variety of tools and methods can you identify the infrastructure as the cause of your troubles.

You can experience three types of basic problems when troubleshooting infrastructure problems:

• Specific failures— A device such as a hub or switch can cease to function and cause an entire section of the network to fail. Such problems tend to be quite easy to troubleshoot. If you are armed with the network documentation, it should be fairly simple to pinpoint the source of the problem.

• Nonspecific failures— Sometimes users experience random problems with the network, but there doesn't seem to be any common thread to the problems. Such problems can be hard to isolate because it is often difficult to pin down the exact cause.

• General performance problems— A common sign of a problem with the network's infrastructure is poor network performance. Of course, poor performance is a vague statement and can be attributed to everything from a misconfigured operating system or application to failed hardware. In the troubleshooting process, all these problems would likely be eliminated as causes before the network's infrastructure would become suspect. You are most likely going to encounter infrastructure problems when using new equipment or new applications on older networks. For example, perhaps the 10Mbps hubs and the Category 3 UTP cable just won't work. It is the network administrator's job to be aware of possible infrastructure problems before installing new network components or applications. You can save yourself a lot of legwork by fully exploring the impact that changes to the network will have on the existing infrastructure before modifying the network.

Baselines and Performance Monitoring

Often, the only way to know if the network is suffering from poor performance is to compare its current performance to its past performance, to see if it has changed. Such a comparison is done by using baselines. Baselines are used to measure network performance and provide a means of comparison when troubleshooting network performance.

Performing baselines on the network is not a one-time task. For baselines to be useful, they must be performed periodically. As a network expands, introducing new users and new applications, baselines can let you know if the network infrastructure can carry the load. As a rough estimate, you should consider the following guidelines in determining the frequency of network baselines:

• Changes to network applications— Some applications are more bandwidth hungry than others. Installing the latest and greatest application might be too much for the existing network infrastructure to handle. To determine whether this is the case, you should perform baselines before and after installing a new application.

• Addition of users— Network performance can slow down if too many users are on the system, using the same resources. If several new users are to be added to a network, a preemptive troubleshooting step would be to obtain baseline performance before and after the new clients are added.

• Changes to the network hardware— Changes such as installing a new NIC in the server or adding an additional hub can have an impact on the network's overall performance.

• Software upgrade— If you have to upgrade your network operating system, you should obtain a baseline. A network that is designed for and working with Linux might struggle if, for example, it were switched to Windows 2000.