August 02, 2017
Fault finding methods
There are literally hundreds of hardware and
software tools that have the ability to help diagnose and solve problems. These
tools are somewhat limited unless you can use them in conjunction with sound
fault-finding methods. This reading will introduce you to fault-finding
methods.
The scientific method to fault-finding
Fault-finding is a skill required in just
about every industry, not just in IT support. If you do a search on the
Internet looking for fault-finding information, you are likely to encounter
massive amounts of literature for many industries, not just IT. In general,
those areas of industry which demands high-level technical skills have
developed well-documented fault-finding methods. It is not surprising though
that all of these methods have similar principles. Enter the scientific method
to fault-finding.
The scientific method is not specific to any technology. The scientific method is an investigative process that uses
logic to test theories or hypotheses through observation and methodical experimentation. In fact, the scientific
method has been around since the beginning of time, when people began to derive
knowledge from the world around them.
The scientific method proposes to use
logical and systematic steps (procedures), to analyse available information,
such as symptoms, in the hope to find information that is useful and relevant
whilst discarding what is not. This procedure will enable you to draw
conclusions and hopefully arrive at the source of the problem. Generally, the
method is repeated (cyclic), until the source of the problem has been
identified.
The principles of the scientific method are
summarised in the following steps:
1.
Gather Information
2.
State the Problem
3.
Form a hypothesis
4.
Test the hypothesis
5.
Draw conclusions
6.
Repeat when necessary
The next part of this reading will introduce
fault-finding techniques, based on the scientific method.
Cyclic fault-finding
Cyclic fault-finding is the preferred method
for problem determination used in the IT industry. The myriad of hardware and
software tools available for fault finding will help you gather useful
information, but generally, the tool won’t fix the problem for you. You will
need to make your own decisions in terms of what is the best course of action.
Generally, companies develop their own
cyclic methods, or choose to adhere to someone else’s method i.e. Cisco’s
Troubleshooting guidelines (see the Cisco website www.cisco.com and search for
"troubleshooting"). The most important part of troubleshooting any
problem is to divide the tasks of problem resolution into a systematic process
of elimination. In general, cyclic fault-finding involves taking a series of
steps, varying from 5 to 8 steps, and then repeating these steps until the
problem is solved.
It is important to note that cyclic method
rely on technicians formulating a hypothesis (probable cause – step 3), and
then testing the hypothesis (steps 4 and 5). If the desired outcome if not
achieved, the process is repeated (new hypothesis). Take a look at the
following steps introduced below:
Figure 1: Steps in Cyclic Fault Finding
Strictly speaking, the above process only
requires 7 steps for troubleshooting, but best-practice is to update/create
appropriate documentation – good quality documentation will only aid
fault-finding in the future!
Fault-finding decision trees
Quite often, IT support companies will
develop fault analysis trees or fault finding decision trees. Fault finding
decision trees are aides for support people to use as guidelines when
troubleshooting.
With time, you will be able to create your
own decision trees to help the fault-finding process. The following is an
example of a decision tree aimed at helping someone troubleshoot a network
access fault (i.e. user cannot log in or access her email).
Figure 2: Sample decision tree to help
someone troubleshoot a network access fault
The steps in the tree above can be explained
as follows:
The tree begins by defining the problem: The
user cannot access e-mail. Then a question is asked (diamond shape). Only two
answers are possible: Yes or No. Depending on the response a process takes
place (rectangular shape), which leads on to another question. No matter how complicated,
all decision trees work in the same manner:
1.
State Problems (Begin)
2.
Ask Questions (Diamond Shape)
3.
Analyse Response (Yes or No)
4.
Take Action (Rectangular shape)
5.
Ask more questions
6.
Analyse response
7.
Take action
8.
And so on until problem is
solved or a different course is deemed necessary
It is not unusual for software and hardware
manufacturers to include such charts with their products, as additional support
information. The general idea is that people may, with the help of these trees,
perform first-level support, potentially cutting down on the number of support
calls made to companies. Effectively, vendor supplied charts become a form of
fault finding tool.
