However in reality, a great majority of the problems we will be asked to resolve will be relating to software, hardware or problems we have never seen before. Our professional illusion is maintained by out ability to properly handle these instances. Below, is a methodology which is common in the industry, and the terminology taken from Adrian Grigorof, B.Sc, MSCE.
The 7 Troubleshooting Techniques
1 Develop fundamental knowledge:
- Initially focus education and training in the fundamentals of computing instead of the specific details of a particular application. Ensure that you have a great horizontal (or broad) understanding of the fundamentals. Unlike a specific corporate IT department which simply needs to focus on 3 or 4 line of business applications, we have dozens we will handle daily. General concepts such as TCP/IP, DNS, DHCP, Authentication, Access Control (ACL, MAC, RBAC, DAC), etc.
2 Understand the problem or what is causing the problem:
- Don’t troubleshoot blind, or by what the user says is the problem. Over half of the problems reported are not repeatable – how can you tell if you fixed a problem if you cannot make it fail? This can be due to user error or intermittent problem (which increases the difficulty for resolution). Ensure that you can both: (a) make the problem occur; and (b) understand what exactly is happening – error codes, timing, processes, axillary data.
3 Do one change at a time:
- Take your time, have a clear goal in mind (Do you know what you’re trying to fix? How will you know if you accomplished it?) Do a single change at a time and test – go slow. If you make multiple changes, you will not be able to determine what caused the problem. Along this note, make sure that nobody else is working on the problem or on the system at the same time you are.
4 Use existing resources and make finding available for future use:
- Use existing knowledge bases, both our corporate KB, along with vendor and vendor neutral sites. When an answer is found, document in the corporate KB for future use.
5 Get and learn tools of the trade:
- A good tool can sole a problem in half the time. There are a vast number of tools available to troubleshoot and find answers. They can also be found on this Wiki.
6 Have a plan, deliver written results:
- Primarily discussed above: Have a plan of attack to resolve the problem and then document the resolution. Be sure to include information in the Service Call record. As appropriate notify the end user or customer of the issue, both verbally and follow-up with an e-mail. When performing services without the end-user or customer contact available, be sure to follow-up via e-mail immediately (i.e. before leaving, send an e-mail from the affected system – even a simple “problem resolved, you should have not future problems, will follow-up tomorrow with details” will be sufficient).
7 Analyze the problem afterwards:
- What policies or procedural failures were there – either on the customer end or our company end. Could this have been avoided and/or what could prevent this sort of failure in the future.