Automated systems and network monitoring systems are great at alerting us to problems. Programs and processes that track your systems’ status work well to help you stay alert of significant issues, but they aren’t perfect. A hybrid approach of automated and manual checks strikes the perfect balance for keeping your systems healthy. An experienced, seasoned technician can often catch errors or anomalies early, which these monitoring systems could overlook until they become more significant problems.
To help detect issues, every time you login to a system, look around to see if you can spot anything abnormal. Create a routine of systematic checks to assess the basic health of a system, and by doing so catch minor errors or symptoms of larger problems. This process won’t take long and should be incorporated into your everyday practices when logging into machines. In 1-2 minutes, you can give a Windows system a quick health check.
As an example of what you can easily do, here’s a timeline of some basic things you can check:
- 0-10 seconds
- Open Explorer (Windows + E) and check for disk space usage. In a default view, Windows provides a nice colored bar for each drive. If it’s showing a red bar for a drive, take a closer look, otherwise move on.
- 10-30 seconds
- Open task Manager (CTRL+SHFT+ESC). Take a quick peek at CPU, Memory, Disk usage, and Network usage. High usage in any of these areas requires additional investigation. If you see normal usage stats, then move on. Your definition of “normal stats” will be based on years of experience looking into system performance.
- 30-60 seconds
- Open services.msc (Windows + R, then type services.msc).
- Sort services by “Startup Type”. Go through all services set to startup automatically. If any are not running, they will need further investigation.
- There are some services set to start automatically that will never be in a running state. In time you’ll figure out what those are and skip over them.
- 60-70 seconds
- Check for failed TCP connections. Open a command prompt (Windows + R, then type cmd). Type in ‘netstat -n 1 | find “SYN”‘ and look for any results. Any connections in a SYN_SENT state indicate the system is trying to connect to something that is not listening and will require additional investigation.
- 70-120 seconds
- Check event viewer for any recent Errors or Warnings. Open Event Viewer (Windows + R, then type eventvwr.msc).
- Expand Custom Views then click on Administrative Events. This log gathers all Errors and Warnings from other event logs.
- After that look at the System and Application logs.
- If the system is a Domain Controller, also check the “Directory Service” event log for any AD-related issues.
With experience, an eye for detail, and basic investigation skills, you can assess a system’s health within two minutes. Creating a health check routine will point you in the direction of probable causes for issues you are looking into and alert you to underlying problems monitoring systems may not be able to pick up on. The small clues you detect can help solve the problems you are fixing and stop potential errors from occurring; creating a routine for system checks when logging into a machine combines reactive and proactive problem-solving. After all, when it comes to troubleshooting, the best way to fix problems is to detect them as early as possible.