4 steps to troubleshooting (almost) any IT issue

Troubleshooting IT can be… tedious (understatement of the year). End users submit seemingly endless problems ranging from complaints of their Internet being “slow” to forgotten passwords to constant printer pains. What’s your plan of action for the next vague phone call, email request about tech issues, or help desk ticket from a big wig who needs his computer fixed… RIGHT NOW?

Worry not! We’ve got several problem-solving steps to follow that’ll help ease the stress of solving nearly any IT issue:

Step 1: What exactly is the problem?

Getting to the bottom of a computer issue can sometimes feel like playing 20 questions, so it’s crucial to ask the right questions first if you want to discover the root problem quickly.

• If you’re talking to someone or emailing with them, listen or read carefully and take notes (mental or otherwise). When they’re done, proceed with questions that either have a yes/no answer or can be answered quantifiably (e.g., “How many times did this happen?”). This approach can move things along and cut down on ambiguous responses.
• What should you ask? The key is to find out the “who, what, and when” of the problem. Does the issue affect just one person or many people? Did it occur right after a significant event on the computer or the network? There are specific questions you can ask to drill down beneath general statements like “I can’t send email.” Are they using a PC or mobile device? Is the PC power on? Are they able to open their e-mail client? Can they send or receive email, or both?
• Once you’ve got a grasp on the issue, try repeating your understanding of the problem so that you’re both on the same page as to what’s wrong. It’s a good idea to use some the same language they used and avoid using complex tech terminolgy. Ideally, this will help remove potential confusion related to the issue.

Now that you have a basic understanding of what the issue is all about, you can look into technical details that could point you toward to a solution.

How can you answer the most common IT questions?

step 2) Gather more details, eliminate variables

In many instances, what was reported as a general issue (e.g., the Internet is down) is actually something very particular, such as a specific website being offline. Best way to find out? Ask those pertinent questions and then dig up more info from various sources such as:

• Error messages: If the user reports one or you’re present to check for one, they can point you in the right direction as to why a failure occurred. For example, Windows blue screens of death provide a relevant error code that’ll give you a good sense of what caused a failure.
• Events logs: Do they exist, and if so, what exactly do they say? In addition to error messages, logs often provide a timestamp so you can answer the question of when exactly events happened. Checking the Windows event viewer can be a first step in pointing you towards any relevant logs.
• Can the user provide screenshots, video, or other supporting information that can help assist in the troubleshooting process?
• Diagnostics results: Have you run system utilities to get even more information? For example: ping can help you remotely check if a server or website in question is reachable. Additionally, the Windows memory diagnostic can check for faulty memory, the resource monitor or performance monitor can check for unusually high CPU or memory usage, and a disk check can examine your hard drives for errors.
• Monitoring: Do you have any 3rd party tools that can offer more clues or predict issues before they occur? A network monitoring solution like GMN Network provides preemptive alerts and warnings about servers at risk of going down. And a network inventory application can provide vital statistics on dozens or hundreds of systems such as disk space, available memory, OS installed, and software running on the devies in question. The event viewer helps you find logs that can point towards the source of a problem

Step 3: Reproduce the problem, develop hypothesis of root cause

Now that you’ve gathered basic background info, it’s time to get hands-on with the problem. Reproducing an issue simply means verifying you can recreate the same error the user reported. You can do this at the physical site of the problem, or through a remote desktop / remote control application. Neither of those options available? You might be able to approximate the same conditions on a similar computer.

Once you encounter the same error(s), you can more easily develop theories as to the root cause based on what you observe, then start taking steps to fix the issue. Sometimes you’ll find the culprit (and come up with the solution) pretty quickly. Other times, it’ll take longer and you may have to dig into knowledge bases, comb through old help desk tickets, or Google the problem for others who have faced the same issue.

At this phase, a deep understanding of how computers work will prove… useful. For example, in computer networking, if you understand 7 layer OSI model for networks (a problem can exist in any one of them), you have a framework for systematically working your way through the potential causes of a networking problem. For a connectivity issue, an Ethernet cable could be damaged or unplugged (Layer 1 issue), network requests might not be going through (Layer 3), or an application might not be properly coded (Layer 6).

Step 4: Attempt a fix based on findings

The evidence you’ve already gathered should have narrowed down possible root causes and positioned you to fix the issue. At this point, focused experimentation is the name of the game. You can try tweaking changing settings related to the problem, swapping out faulty parts, repairing corrupted files, updating drivers and software, etc. until you solve the problem (or at least get closer).

Still stumped? You can always restore the machine back to a point when everything was working fine. That is, if someone took the all-important step of backing up data or system state before the problem happened so you can go back to a simpler time… before problems plagued their PC.

While every issue is like a unique snowflake, a significant number of issues can be resolved through common troubleshooting steps like rebooting the problematic machine, checking for DNS and DHCP issues, checking the device manager for driver issues, cleaning up a machine, or checking firewall or proxy settings, etc. For issues that are tougher, if your own experimentation and Google searches don’t result in a satisfactory result, browsing discussions on IT forums such as GMN Network  can be extremely helpful for fixing PC issues. If nothing turns up there, you can always ask the community of millions of IT pros in GMN Network.