Diagnose network problems

Typical network problems — hardware

Typical network configuration problems

Main configuration settings

Internet Protocol (IP) overview

Common configurations

Settings on the computer

Example: Settings with Microsoft Windows XP operating system

Testing the settings on the computer

Testing settings from another computer

Typical network problem-solving techniques

Summary

Check your progress

Typical network problems — hardware

Resolving common hardware problems often involves first checking cable connections are securely in place. Ensure cables are placed where they cannot be nudged out of sockets, stepped on or have office furniture or chairs placed on them (which can damage the fine wires inside network cables particularly).

For network problems, check that all the cables for connection devices such as hubs and switches are inserted correctly into sockets.

Cable plugs don’t have to come out completely for a bad connection to exist. The network cable plugs RJ45, for instance, have a small catch that holds the plug in place. If this catch is broken off, the plug can lift slightly out of the socket, which is enough to break the connection.

Bad cables can also be a deadly problem. For example, wrongly wired UTP cables may test correctly using a static connectivity tester, but will not run at Ethernet speeds due to incorrectly configured pairs. Also check that a crossover cable is not being used. These cables have the wiring changed at one end and are used to connect two computers directly together. They may have been introduced into the company from a small office and may not be labelled.

Untidy cables can tangle with their weight prising connections loose. Connections might also loosen if cables are in an open space where people brush past or other hardware items pull on them.

Cables should be securely fitted around the walls. They should only be safely run across the floor in the low profile, usually hard rubber channels, specifically made for that purpose. Fitting cables properly not only ensures continuous network connection but also is a requirement of Occupational Health and Safety regulations, to prevent accidents (from people tripping over cables).

Typical network configuration problems

If a computer cannot view other computers on a network, or cannot access resources such as printers, then network configurations can be the problem. To check a computer’s configuration to the network you can use the commands described below. Some companies use shadow software to assist technical staff. With shadow software installed on client computers, the technician can ‘take over’ the client computer from their own workstation, run any tests, and make the necessary changes to configurations.

Main configuration settings

To join a Local Area Network (LAN), computers must usually have the following configurations in network settings:

  • Network card driver installed
  • IP address
  • Subnet mask
  • Network client software
  • Name of the workgroup or domain of which the computer is a member.

If you want others to access files on your computer, or if your computer acts as a host to a printer, then you will also install ‘file and print service’.

If the network is configured as a client/server network or domain, then each computer must have:

  • a Domain Name Server (DNS) address
  • Dynamic Host Configuration Protocol (DHCP) enabled (if the IP Address is to be obtained automatically when the computer is turned on).

If the network is configured so client computers can access another network or the Internet, each computer may need to be configured with the default gateway IP address. Also, the browser used to view websites must have its connections correctly set, which may include proxy server IP address and port settings.

Internet Protocol (IP) overview

Internet Protocol (IP) is the main protocol used to connect networks. The most common IP addressing system currently used is IPv4, which we will look at here. This is a 32-bit addressing system. A newer and more encompassing addressing system, IPv6, which comprises 128 bits, is planned to supersede Ipv4 in time.

To arrive at 32 bits, the IP address is broken into 4 octets that consist of 8 bits each. A bit is a 0 or a 1 and each octet is changed into a decimal equivalent number between 0 and 255.

There are three classes of IP addresses, and each of them has a network portion and a host portion. A subnet mask helps distinguish the two parts.

Table 1: The three classes of IP address

IP address / Range / Subnet mask
Class A / 0.0.0.0 to 127.255.255.255 / 255.0.0.0
Class B / 128.0.0.0 to 191.255.255.255 / 255.255.0.0
Class C / 192.0.0.0 to 223.255.255.255 / 255.255.255.0

There are a number of IP addresses that cannot be used on networks. They are used for special purposes such as Internet testing, local testing, or can only be used for LANs.

  • In a Class A network the first octet, being the network portion, must be the same on each computer, and the following three octets, the host portion, must be unique to each computer.
  • In a Class B network, the first two octets must be the same on each computer and the host portion (being the last two octets) must be unique to each computer.
  • In a Class C network, the first three octets must be the same on each computer for the network portion, and the last octet must be unique as this is the host portion.

A host computer can have an IP address set as a static number. This means they will always have the same IP address. On larger networks, DHCP is used for the client computers. This means that each time the computer is turned on it is given an IP address and other configuration information from the DHCP server. In this case, the DHCP server is software that is installed, often in the main login server, such as the domain controller.

Common configurations

A common IP address used in many private networks is 192.168.0.x, where x is the unique number for each host on the network. A private network is not part of the Internet. A host is usually a computer or a networked printer. Often the first few numbers on a network are reserved for servers. For example:

192.168.0.1Main login server

192.168.0.2Backup login server

192.168.0.3Database server

192.168.0.4Web server

192.168.0.5File and data server

192.168.0.6 to 192.168.0.9Reserved for new servers

192.168.0.10 to 192.168.254Clients, ie. individual computers.

If there are only a few computers in a network, each one may have the setting configured manually. This is known as a static setting. In larger networks, a DHCP server will issue each of the computers an IP address as they are turned on. This is known as a dynamic setting.

Settings on the computer

As you can see, there are many places where configuration problems can occur. To check these, you can view each of the configuration settings on a computer, or you can test it from another location.

Example: Settings with Microsoft Windows XP operating system

Here we will consider the common settings on a client computer with the Microsoft Windows XP operating system. There will be other variations if you are using other versions of Windows operating systems, or are in a Linux environment, or you use an Apple computer, or if you connect to other servers such as Netware or Unix. To make these configuration changes you may need the operating system disk.

Local area connection properties

You access local area connection properties from the computer desktop. For example, select Start, Connect to, Show All Connections, then right-click on Local Area Connection and choose Properties.

To install File and Print sharing, just tick the box. To install the ‘Client for Microsoft Networks’ you choose Install, then Client, and then follow the prompts. To install TCP/IP, choose Install, then Protocol, then TCP/IP and follow the prompts.


Figure 1: Window to adjust the TCP/IP setting or view it after protocol is installed /
Figure 2: Screen to adjust the default gateway and DNS settings

To adjust the TCP/IP setting or view it after the protocol has been installed, you access Local Area Connection Properties, highlight Internet Protocol (TCP/IP) then choose Properties, as in Figure 2.


Figure 3: System properties screen showing tab for computer name / From System Properties you can also adjust the default gateway and DNS settings. From the Advanced tab you can edit the IP settings, add further DNS settings or WINS settings. WINS is a protocol used to resolve names to IP addresses (like DNS) for older versions of Windows operating systems.
Once the various settings are configured correctly, you can then check that the computer has its membership correctly set in the workgroup or domain. Choose Start, then Control Panel, then System. From this dialogue box choose the Computer Name tab, and then click Change.
The name of each computer or host must be unique. Type in the name of the domain or workgroup that the computer needs to join.

Testing the settings on the computer

As well as going through the screens so far mentioned, you can also check the connection using commands. One method to access the command prompt is to choose Start, Run and enter ‘cmd’ in the Open box.

The command ping 127.0.0.1 is known as pinging the loopback address. This indicates that the computer is correctly configured to the network card.

Figure 4: Command prompt screen for information from ping 127.0.0.1

The command ipconfig displays the minimal information on the configuration settings on the computer. In the following example you can see that the IP Address is 192.168.0.1 and the subnet mask is 255.255.255.0.

Figure 5: Command prompt screen for information from ‘ipconfig’

For more information you can use the switch/all. Type the command ipconfig/all.

Figure 6: Command prompt screen for information from ‘ipconfig/all’

Testing settings from another computer

You can run a number of tests from your workstation. This can help avoid delays in moving to a disconnected computer and disrupting the user.

The ping command is one of the most useful checks of basic connectivity between you and a target IP address. It sends out an echo request, the target system receives it and then sends back a response. If you receive a response, it means that there is an active computer or device at the address. You also get the time (distance) taken to reach the target, and back. The command is ping followed by the IP address of the target computer, or ping followed by the name of the computer. For example, ping192.168.0.1 or pingSales1 or pingmix as in Figure 7 below.

Figure 7: Command prompt screen for ping followed by the IP address

If the computer you pinged is not on the network, you will receive a ‘timed out’ reply. This could mean further investigation on the computer is required to adjust configurations. If you ping by name, and get the first line showing that the system has resolved the name to an IP address but then ‘timed out’, you know that you have connection to the network, but there may be a problem with that workstation. There are various switches you can also use with the ping command. See Table 2 for further examples.

Figure 8: Command prompt screen for information from ‘ping 192.168.0.5’

Tracert command

The tracert command lists the number of hops needed to connect to a target computer. Each hop represents a device that the packets of data must travel through in order to continue along the network. On the Internet, this is a usually a router. When testing in a LAN environment, there is usually just one hop. The example in Figure 9 is from testing a computer in a LAN.

Figure 9: Testing a computer in a LAN using tracert

Figure 10 shows a tracert command used when testing over the Internet. You can either use the name of the website or the IP address, if you know it. The following example looks at the number of hops it takes to reach the Australian Taxation Office website.

Figure 10: Testing a computer on the Internet using tracert

Table 2: Summary of commands to test network connections

Syntax and example / Description
ping IP_address
Example ping 172.16.157.200 / Sends out 4 echo requests to the target IP address and displays the results.
ping –t IP_address
Example ping –t 172.16.157.200 / Sends out echo requests every second until you press ‘Ctrl-Break’ or ‘Ctrl-C’. Very useful for tracking down network problems when there is no network activity.
ping 127.0.0.1
Example ping 127.0.0.1 / This is a special IP address that will display the result for the local computer or device.
tracert IP_address
Example tracert 172.16.157.200 / Using tracert to trace the route within a single LAN. This gives similar results to the ping command.
tracert IP_address
Example tracert 224.44.168.13 / Using tracert to trace the intermediate routers and gateways gives similar results to ping for each of these devices or systems.

Reflection activity

Look around a workplace to which you have access.

  • Do certain kinds of network problems regularly occur?
  • How are the problems solved?
  • What networking activities take place?

Typical network problem-solving techniques

In any troubleshooting procedure a series of tests narrows down the possible causes of a problem. The following procedure can be used as a guideline — each organisation will have their own procedures to suit their computing environment.

Table 3: When the user cannot access a peripheral on the network — typical techniques for troubleshooting

Item / Commands / 
Check that the network cables are securely in place at the back of the computer. / 
Check that cables to hubs, switches, bridges and routers are fixed securely in their sockets. / 
Check the network configuration on workstations involved, to ensure that they are correct. / Use the commands ‘ipconfig/all’ and ‘ping’. / 
Check that you can access other computers and peripherals on the network. / Use the commands ‘ping’ or ‘tracert’ if part of your network is on the other side of a router. / 
Check the permissions to ensure that the user has the right levels of access network resources. / 
Check that an individual user is in the correct group that has access to network resources / 
Check and replace the network cards and drivers, as required. / 
Check the configurations of the default gateway and browser proxy server setting. / Use the command ‘ipconfig/all’ for the default gateway, and ‘Tools’, ‘Internet Options’ in the browser to check proxy settings. / 

Reflection activity

Look around a workplace to which you have access. What procedures are in place to solve network connection problems? Is there any formal written documentation to follow when a problem occurs?

Summary

In this reading you’ve considered typical network problems. Disconnected cabling can often cause problems. Once hardware has been checked it usual to check the configuration of the networking software, and settings such as the IP address, subnet mask, DHCP configuration, DNS and default gateway IP addresses.

The standard procedure to follow in order to fix network problems efficiently was outlined, which, in summary, is to check hardware connections, network configurations and testing.

Check your progress

Now you should attempt the Practice activities in this topic. If you’ve already tried them, have another go and see if you can improve your responses.

When you feel ready, try the ‘Check your understanding’ activity in the Preview section of this topic. This will help you decide if you are ready for assessment.

Reading: Diagnose network problems1

2005