Random crashes are annoying and difficult to diagnose. I'll review common causes and steps you can take to determine what's happening on your computer.
A “crash”, for the purposes of this discussion, includes things like random blue screens, random reboots, freezing up and just randomly shutting down or going to a completely black screen without warning.
I actually have several articles on crashing randomly, but because this is such a common scenario I want to update my approach a little.
In particular, I find one cause to be the most common.
I’ll look at that, and then review what else might be going on.
The most common cause of truly random crashes is an overheating computer.
The most common cause of an overheating computer is blocked airflow through the computer, followed by a failing or failed fan.
You probably know that there are fans in your computer whose job it is to move cooler air throughout so as to cool down the components therein. Your computer’s processor, hard disk and other components all generate heat, and if that heat is not somehow dispersed, the components themselves become too hot, function improperly, and cause a crash.
Particularly when it comes to laptops, it’s very easy to accidentally block the air vents that allow air to flow properly through the device. I actually recommend never putting a laptop on your lap: use a table or “laptop desk” type device that ensures proper airflow.
Desktop computers are more vulnerable to dust and pet hair which can clog the air vents. It’s worth unplugging your computer and looking inside. Carefully clean it out if you find large quantities of dust and dirt in there.
Fans do fail. Fans have actually become slightly more complex over the years – in a effort to reduce the noise generated by your computer fans are now often variable speed – running quietly at a slow speed when your computer is relatively cool, and speeding up to move more cool air through more quickly as the computer heats up.
Heat and your CPU
The temperature inside your computer can change depending on what you’re using it for.
The harder the computer’s processor or CPU works, the more heat it generates. That’s actually what makes those variable speed fans make sense – if your computer is mostly idling then it’s quite likely you don’t need the fans to run full speed. On the other hand, if the CPU is working hard, then as the internal temperature rises the fans will speed up to cool things back down.
Well, they’re supposed to.
Listen to, or watch your computers fans as you use your computer. If they never turn on then there’s almost certainly a problem with the fan or its control circuitry.
To be clear: your computer should be designed to be able to handle the maximum heat the CPU can generate. Even if your CPU is pegged at 100% usage in Task Manager, the fans should keep up. A machine that starts crashing randomly the longer it is used, or the harder its CPU is put to work is giving you a warning that something about the cooling system isn’t working properly.
Check for overheating first. After that it gets more difficult.
I used to mention malware early in the discussion about random crashes, but these days malware is much more interested in keeping your machine running so that it can do whatever malicious deed malware does. Certainly look for malware, but it’s less commonly a cause of random crashes than it once was.
Failure is always an option. Specifically hardware can fail. We commonly think of hardware failure as being instant, total and catastrophic, but in fact there are various components that can have intermittent failures, or whose failures can have a delayed or random effect. Next to the fan, which we’ve already discussed, bad or failing RAM is by far the most common hardware-related cause of random crashes. Programs don’t always get loaded into the exact same place in memory, and that means that you might sometimes access bad RAM immediately, causing a crash, and other times it might take a while.
Software tends to be what people think about first, but in fact with current Windows versions the vast majority of software errors will not result in your computer actually crashing entirely. More commonly, a program will crash or terminate unexpectedly, but Windows itself will keep on running. The exception to this is device drivers. Device drivers often operate at a low level where a bug or other unexpected error could cause the driver to fail in such a way that the entire system would crash. In particular, if you’ve recently added hardware to your system that involved some additional drivers being installed, or if a recent update included device driver software, then it belongs on the short list of suspects potentially causing your random crashes.
The hard disk is a bit of hardware that deserves special mention, because it can mess up the software. If your hard disk, for example, has bad or damaged information on it (bad sectors, for example), and that damaged information happens to affect a part of Windows, or a device driver or something else, then when your system attempts to use that bad information a crash might result. Typically, you’ll get signs that this is coming: significant slowdowns in disk access, or even “bad sector” or “crc” warnings, before the crash.
When faced with random crashes, here’s a plan of attack:
- Make sure your computer has unrestricted airflow.
- Make sure your computer is free of dirt and dust, and that the fans are all working properly.
- Run a CPU temperature monitoring tool. (RealTemp appears to be one viable utility.)
- Run a memory diagnostic such as Memtest86.
- Run a hard disk diagnostic such as SpinRite.
- Make sure that the operating system and all device drivers are as up-to-date as possible.
- Make sure that anti-virus and anti-spyware utilities are running and up-to-date.
And as always, if the crashing behavior started after a recent change, consider reverting to a system backup taken prior to that change to see if perhaps that change was the reason.