The Basics Of Network Troubleshooting

Overview

The process of troubleshooting your network involves a methodology that starts with cabling and works through the OSI model to the application layer. The network devices have a network cable that terminates at a wired switch. Cabling is a source of a lot of network problems.

The key with troubleshooting is to determine what has changed. Sometimes it is hardware that isn't working or some change was made such as new software, configured equipment or additional employees stressing the network. It could be a specific issue or several problems. Start with the client adapter, determine the specific symptoms and go from there.

The following is a series of questions that can used as a guideline during the troubleshooting process:

1. Can you ping the switch IP address?

2. Can you ping the router?

3. Can you ping the DHCP server?

4. Does the client obtain an IP address?

5. Is the WAN circuit available?

6. Has new software been installed on the client or server?

7. Do all clients experience the problem?

8. Is it random or a specific pattern?

9. Is the problem server specific or application specific?

10. Is the DNS server IP address setting correct?

11. Is there a firewall that is filtering traffic?

12. Ran traceroute and examined routing behaviour?

TOOLS:

1) Windows control panel shows network adapter settings, firewall configuration etc.

2) Ping and traceroute will verify that network routing is working.

3) Examine the ARP table on the desktop, switch and router confirming the device MAC address is there.

4) Network packet sniffers examine packet information such as protocols, filtered programs or errors with applications.

5) Verify change management activities and determine what if any device was changed and how.

6) Show interfaces at the router will reveal any interface errors pointing to a cable or hardware problem.

7) Examine router interface utilization patterns.

Working from the cable level and determining if the problem is affecting a single user, a department, building or city wide is a good place to start. A city wide problem sometimes indicates a data center outage. A building sometimes points to a circuit, router or primary switch. A department can indicate a problem with their network switch or fiber cabling. The single user problem could be a myriad of issues from a bad cable to network permissions. Because companies have in some cases hundreds of applications, network switches and routers the problem becomes more complex when it is a cross department issue. Not all employees have the same network permissions and use a variety of applications across departments, cities and countries.

Determining who is affected by the problem can be a key factor in resolving the issue. The ping is a popular tool since it verifies to the network layer all is working. In that case your then focusing on application issues although not always. Traceroute verifies how packets are traveling between source and destination. You could have packets using non optimal routing paths that are causing performance problems.

Shaun Hummel is an author of various technical books and has a web site focused on information technology job search solutions and certifications. www.networkjobsolutions.com

» posted by networkjobsolutions

Sign up for ITworld's Daily newsletter
Follow ITworld on Twitter @IT_world

• Email
• Print
• Share
  • Slashdot
  • Digg
  • Stumble
  • Reddit
  • Newsvine