Troubleshooting PC Problems
Troubleshooting Problems
As an IT technician, troubleshooting PC problems will be
one of your core job responsibilities. Troubleshooting is both art and science
and you can learn how to do it. Troubleshooting also happens to be why many of
us sign up for jobs in IT and enjoy the day to day – troubleshooting and
diagnosing problems excites us!
Essential Tools for Troubleshooting
There are many tools you can use to work on PCs, but there
are two essential tools you should bring when troubleshooting a problem – pen
& paper and an ESD strap. You will learn in the Safety section how crucial an
ESD strap is, but it can mean the difference between a fixable computer and a
dead piece of junk.
Pen & paper are crucial to troubleshooting so you can write
things down you will need to remember later – such as error codes, software
settings, or a web address. You don’t want to have to bother every customer
you visit for a pen and paper so you can write things down. You also do not
want to rely on your memory for an error code. A typically Microsoft error
code might read: 0x878225xCCC00 LOAD SICILY FAILED. Will you remember that on
the trip back to the office or upstairs to your machine?
Essentially, there are three reasons why you will be called
to troubleshoot a PC problem:
1. User error – also
known as training issue – the user is trying to do something he or she does
not understand or is doing incorrectly and thinks the computer is “broken.”
2. Recent changes to
software or hardware.
3. Actual computer
hardware failure.
The third reason is actually a much smaller percentage of
the problems you will face than the first two. In troubleshooting, you need to
determine first if the problem is just user error or if something changed on
the user’s computer.
A change on the computer could be an
intentional change (I installed this great piece of software off the
Internet which downloads cute pictures of kittens, oh and also broken my
accounting application) or an unattended change
(Windows update installed a security fix which broke the accounting
application).
In a small percentage of service calls, you will be lucky
enough to have an actual piece of broken hardware to fix!
To figure out what kind of event happened to cause the
service call, you will need to ask the user several questions.
Diagnostic Procedures
Communicate
The first step in diagnosing a problem is to
communicate with the user. Determine exactly
what the user believes the problem is – this will go a long way in determining
the true issue and how to best resolve it. The user may describe a symptom to
the problem and not the actual problem. For example, a user might say:
“I’ve got a 6:00pm flight to catch and this stupid printer
won’t spit out my document. I need this document before I go and I need to
go!”
The real issue is that the user needs the document printer
so she can run out the door. Spending several hours troubleshooting why the
particular printer/document won’t print won’t help her – instead, getting the
document printed and then getting the overall problem
fixed is the right order – get her the document so she can get out the door,
then fix the root problem!
Alternatively, the user might say something like this:
“I can’t get my email!”
When really, the problem is the user
isn’t even connected to the network. By identifying what the user
believes the problem is also helps in one other way – it puts some constraints
on what you have to do so you do not get stuck fixing every possible thing on
his machine without solving the one thing he wants you to fix.
You should ask the user probing questions to determine the
scope of the problem:
· When did the problem
start?
· How often does the
problem occur?
· Do you remember
adding any software or hardware before the problem occurred?
· Can you re-create
the problem for me?
· Does it happen in
certain applications or at certain times?
These questions can help narrow down what the problem is.
Backup the Computer
It is good practice (and a question on the exam!) to backup
the computer prior to making any changes during troubleshooting and repair.
Before you identify what you think the problem could be, backup the computer
using Windows system backup or your company’s preferred backup software.
Recognize and Isolate Issues
Recognizing and isolating the issues is the essence of
troubleshooting. You should start troubleshooting wide – not ignoring the
obvious – and move in closer to the problem. For example, let’s look at a
situation where you are troubleshooting a user who cannot connect to the
network:
In this example, we start at the general problem, “No
Network Connectivity”, and then increasingly get more detailed throughout
troubleshooting. We may have gone in a different direction at one point, but
eventually, we troubleshoot enough potential problems to find the core
problem.
Real World
Example
There was once a situation where we were troubleshooting a
network connectivity issue. A user couldn’t connect to the network. We went
through a troubleshooting process of identify the problem from a “large
circle” perspective and determined the problem could be in any of these areas:
- Network Issue (e.g. bad switch, bad wall port, DHCP
server down)
- Network Cabling Issue (bad patch cable, bad cable on the
back end)
- PC Network Interface Card (bad card, bad port, wrong
card settings)
- Software Issue (bad driver, incorrect configuration
settings, Windows issue)
We worked on this issue for awhile and decided the user had
a bad cable and replaced it. The problem still occurred and we searched for
another solution. Several troubleshooting steps later and we came back around
to it being a cable issue. Again, we replaced the new
cable with no resolution to the problem. After much more troubleshooting, we
replaced the cable a third time and the problem was
resolved!
We had a bad original cable and two bad replacement cables.
No matter how obvious you think the solution is – and think it can’t be that
simple – try it anyways! You never know when the solution isn’t just sitting
in front of you. I often find in troubleshooting that the most simple
explanation is usual the right one.
Setting Priorities When Troubleshooting
When you are troubleshooting a problem, you must learn the
art of setting priorities. You often don’t have time to try every
troubleshooting path you think of, you must learn how to decide the most
likely cause, go down that troubleshooting route, and return if you need to.
Much of this ability comes with time and experience – the most you experience,
the easier it becomes to jump to a conclusion as to the most likely problem.
One of the methods you can use to set priorities is by making a list of the
likely causes and then ranking them 1 to 4 with 1 being “Most Likely” to 4
being “Least Likely.”
For example, let’s assume a user is having a problem with a
“slow computer.” If there isn’t a more vague complaint! There are a lot of
causes for “slow computer,” so begin by making a list and marking each item
with a 1 to 4 rating:
1. Most Likely Cause
2. Likely Cause
3. Somewhat Likely Cause
4. Least Likely Cause
My list might be something like this:
Adware/spyware software (1)
Bad motherboard (4)
Degraded hard drive (2)
Too much software installed (3)
Service Pack installation failure (4)
Needs additional memory (1)
You could probably come up with another 20 items, but
that’s a good start. Now that I have a couple theories on what the problem
could be, I will begin troubleshooting the computer by checking for spyware,
checking memory usage, examining the hard drive for fragmentation, checking to
see what software is installed, and so on. The list gives me ideas as to
what the problem might be, and provides ideas on what
order to attack the problem.
When You Hit a Wall
What do you do when you hit a wall in troubleshooting?
One of my favorite scenario questions when interviewing job
candidates is a troubleshooting question with no clear answer. The question is
designed to gage the technical depth of the candidate – push them to their
technical limit – then find out what they do when they hit the proverbial
wall. So what do you do when you hit a wall in troubleshooting and need to
find the solution?
Find Someone Senior
The first thing you might try is to ask someone senior to
you if they have seen this problem before. This step is designed for a “quick
hit” – if they have not seen the problem, you should not just hand it off to
them to solve. It’s a learning opportunity for you to troubleshoot and
research a problem to completion.
Web Research
The next step is to research the issue using the Internet.
There are hundreds of good knowledge bases and forums out there for you to use
in research, but quite frankly, it comes down to Google.com and Microsoft.com.
These two resources will help you find the answer to almost any problem.
Actually, in my many years in IT, I have only had
one problem which we were unable to use the
Internet to find a solution. We found one user in Germany who had a similar
problem to ours and no answer on the forum. Unfortunately, he didn’t reply to
our emails so we ended up troubleshooting the problem for
weeks until we found a solution.
Otherwise, the Google has always been the best resource to
find a solution.
It is often the case that how you phrase your search term
will depend on if you find the answer to your problem or not. You might try
several different versions of your search term to see what results come up.
I have found that Microsoft’s internal search engine is
lacking. To find something in Microsoft’s knowledge base, try this search
term:
Site:Microsoft.com inurl:kb my search phrase
You should also be aware that many hardware vendors have
extensive knowledge bases and forums on their sites. Some are better than
others, but all have great information to help track down your problem.
Hardware Troubleshooting
If the problem is related to hardware, there may be some
obvious clues as to what the issue is. For example, if the computer is beeping
during POST, you are dealing with a hardware issue. For the A+ certification
exam, you should be aware of how to troubleshoot hardware issues.
BIOS POST Diagnostic Beeps
During Power-On Self Test (POST), the BIOS checks different
hardware to ensure it is operating correctly before starting the bootup
process. This POST check will produce several beeps if something is not
operating correctly. Here are several common beep codes and what they mean:
0 Beeps:
Power issue or
problem with the power supply.
1 Beep:
If at the end of
POST and the computer boots up, no problems.
2, 3 Beeps: RAM issue –
reseat memory or replace with known good memory chips.
4, 5, 7, 10 Beeps: The
motherboard has a serious problem and should be repaired/replaced.
6 Beeps: Keyboard error.
8 Beeps: Video card error.
Reseat and check connections.
9 Beeps: Faulty BIOS.
Replace the motherboard.
These are guidelines to typical beep code errors. You
should always check your specific computer manufacturer’s documentation for
the exact error message. POST beep error codes are often a series of long and
short beeps with specific meaning to help in troubleshooting.
Exam Moment
One of the frequently asked questions on the exam involves
a computer system continually losing time setting. This is often caused by a
CMOS battery which has lost its charge. Replace the CMOS battery and the
computer will start retaining its time settings.
Document Your Findings
As you troubleshoot, you should keep written record of your
progress. This will help other technicians assigned your case if you are off
work, and also help you narrow down what the problem could be so you do not
repeat troubleshooting you have already performed.
At the end of your troubleshooting and once you have solved
the problem, you need to document your findings in whatever trouble ticket or
knowledgebase system your company uses. This ensures that the next technician
who picks up a problem for this user or finds a computer with a similar
problem will have access to what you did to resolve the issue. This will save
valuable time and money in the future. |
