What should you do when you suspect that there are multiple issues to troubleshoot?

Page history last edited by jcouchma@... 3 years, 11 months ago

Network Management

1. Network Troubleshooting
   1. Troubleshooting Principles
      1. Gathering information 
      2. Analyzing information 
      3. Eliminating possible causes 
      4. Formulating/proposing a hypothesis 
      5. Testing the hypothesis
   2. Structured Troubleshooting Approaches
      1. Top-Down 
      2. Bottom-Up 
      3. Divide-and-conquer 
      4. Follow-the-path 
      5. Spot-the-differences 
      6. Move-the-problem
   3. Networking Troubleshooting Tutorial
   4. Saving Time with Troubleshooting

Network Troubleshooting

Troubleshooting Principles

1. Gathering information 

2. Analyzing information 

3. Eliminating possible causes 

4. Formulating/proposing a hypothesis 

5. Testing the hypothesis

Structured Troubleshooting Approaches

Troubleshooting can be an extremely difficult process if you don't know where to look. Various network issues may present themselves in ways which don't always make the solution obvious. However, using a structured approach can reduce down time, and mitigate any headaches associated with figuring out what could be causing any issue(s) within your network.

The following structured approaches each have pros and cons; however, each one may be best suited for specific issues, or a network administrators level of expertise when diving into these types of problems. If you have a general idea where the issue may exist, that can also influence which approach you use.

1. Top-Down 

1. To understand the first three troubleshooting approaches, we must first understand what the OSI model is, and how it can be used to make the troubleshooting process easier.

2.  

3. As depicted by the image above, the OSI model is comprised of 7 layers. The top-down approach (as its name implies) starts with the topmost layer of the OSI model (Layer 7) and works its way down until we get to the bottom layer (Layer 1). Each layer depends on all the layers underneath it, so if the application layer is functioning properly, then layers 1-6 must also be fine. This approach works well if the problem is application related, because then your search stops at layer 7, and you can begin solving the issue from there.

2. Bottom-Up 

1. The bottom-up approach begins with the bottom layer of the OSI model (layer 1) and works its way up through the various layers until layer 7 is reached. This approach works well if the problem is likely physical in nature (i.e. a bad cable or a loose wire). Once a physical problem is solved your search will likely end there (unless multiple issues exist).
2. However, this approach requires an intensive look at each layer as you progress up through the OSI model.  For example, you have to check every interface and cable before you can say for certain that the physical layer isn't the problem. It can be extremely time consuming, but is often the favored approach when the root cause of a problem is unknown.
3. Ping and Tracert are useful tools when investigating connectivity issues. If a ping fails this typically indicates the issue exists within the lower layers (1-3), so the bottom-up approach would be the most effect method to use in these scenarios. 

3. Divide-and-conquer 

1. This method is one of the most common. You would choose the layer where you suspect an issue exists and, based on the results of your testing, move up or down in the OSI model accordingly. This method can save you a lot of time if your starting layer is close to the problem; however, if your initial instincts are wrong it could take longer to find the root cause.

2. If a layer checks out, then you know all the layers under it must be fine. Alternatively, if a layer fails your checks then the problem likely exists in one of the lower layers. When using this method it is the most common to begin your testing in layers 3-5 since they are in the middle. Then based on your finding you can react properly to bring yourself one step closer to a solution.

4. Follow-the-path 

1. This method is one of the more basic approaches and is usually complemented by the top-down or bottom-up approach. You basically discover the exact path the traffic takes and examine this route to eliminate unnecessary devices and only focus on the hardware/software which is responsible for the traffic moving from point A to point B. Once this path is determined the issue likely exists someone on one of these devices. 

2. This method will rather quickly lead you to the problem area, and narrow down your search to a particular device. Once that is discovered you can utilize one of the previous troubleshooting methods to find the root cause of the problem. 

5. Spot-the-differences 

1. Also known as the "Compare-Configurations" approach. As its name implies, you would take the setup/configuration of a working environment and compare it to your non-working setup. This will allow you to spot any configuration errors or typos between the various devices. If a discrepancy is spotted, you have likely found the source of the issue.
2. This approach works best in brand new environment that has just been setup. In this type of situation you are more likely to make a configuration error which results in a non-working environment. After major changes are made to a network infrastructure you can also benefit from this approach; however, if you know for a fact that no changes have been made then this approach will likely yield no results.

6. Move-the-problem

1. Also known as the "Swap-Components" approach. This method is useful if you suspect a faulty component may be the issue. Simply take the device/cable/component and move it to another similar system that is known to be functional. If the problem follows the component, then you have discovered the error. If the issue stays where it was, then something else is to blame.
2. This method will usually only be attempted if one of the other approaches point towards the issue being related to a physical component. You will likely only attempt this type of troubleshooting is all else fails. Additionally, not all components are capable of being moved. You should take this into consideration before attempting this method.

Networking Troubleshooting Tutorial

Saving Time with Troubleshooting

When it comes to troubleshooting, it is incredibly important to make sure any potential network issues are taken care of in a timely matter. This will help prevent other problems from arising and/or catching them in the act. That is why there are helpful monitoring tools and software out there nowadays to help the networking processes.

What is the first step you should take to troubleshoot the issue?

What process do you use when troubleshooting IT issues?

What are the 5 steps to troubleshooting?

What are the six steps you should follow to troubleshoot?