Hard disk errors can cause immense damage. But you can take precautions and, if in doubt, correct them – if you notice. We explain what you can do to prevent errors and what to do if the hard disk has a defect. However, it is always advisable to make a backup of the data before the hard disk gets a failure. How to make a backup with Windows is explained here.
Different types of errors
First of all, it should be clear what is meant by “hard disk” here. For one thing, of course, this is simply the physical hard disk itself. All kinds of persistent and sometimes expensive defects can accumulate here. On the other hand, this is also the file system of the hard disk, i.e. under Windows mostly NTFS and for mobile data media often FAT32 or EXFAT. Of course these are not a fixed part of the hardware, but as far as unreadable files are concerned, they can be just as mean.
Besides, it makes a huge difference whether you use an HDD or an SSD. Hard Disk Drives (HDD) are the old, large magnetic data carriers, usually in the order of several terabytes. The data is read in a similar way to a record player: A pickup runs over rotating disks, meaning there are lots of mechanics and moving parts. Solid State Disks (SSD) are flash memories, roughly speaking giant USB sticks – and consequently no moving parts. The possibilities for errors and repairs are correspondingly different.
The actual detection of errors, however, is not entirely trivial. In general, you should suspect hardware and file system if files disappear or the contents of files are faulty, if files cannot be read or written to, or if data carriers are not recognized by Windows from time to time or simply thrown out of the system. If the Windows system itself hangs regularly, this could also be an indication (though it could also be an indication of many other things). Especially with HDDs there is still an unmistakable sign: The Click of Death. When the record is no longer readable and it keeps clicking loud and clear, the pickup hangs – similar to a turntable pickup. In general, loud, irregular noises are not a good sign. An even and slightly “wafting” hum, on the other hand, need not worry you. But it is better than interpreting defects, to see them coming before they occur:
Monitor hard drives
Both HDDs and SSDs have a system called S.M.A.R.T.: Self-Monitoring, Analysis and Reporting Technology. This system transmits all kinds of status information and data about the state of the HDD. A good tool for the Windows desktop is CrystalDiskInfo. The free program shows the temperature, runtime, serial number, connection type and lots of details about the health status via a simple status display.
This is useful in two ways: On the one hand, you can use all this information to ask for help, for example, from the support team. On the other hand, you can see at a glance if any of the installed hard drives are not running perfectly. As soon as a warning message appears here, you should urgently investigate the corresponding point on the net. The good thing about the system is that S.M.A.R.T. and CrystalDiskInfo do not start complaining when it is too late. Usually you still have time to fix the problem.
Repair file system
Under Windows there has always been the tool “Checkdisk”, or ChkDsk for short. The tool can correct bad sectors, identify lost allocation units, cross linked files and directory errors. The last three errors are problems with the file system – and they are not very exciting. However, bad sectors are actually hardware defects that ChkDsk can “repair”: More precisely, the bad sectors are simply marked as “bad” and will not be used in the future. By the way, if there are too many defective sectors, S.M.A.R.T. also issues a warning.
You can use the tool in two ways: Graphically and in the terminal. On the desktop, call up the properties of the suspicious hard disk via the context menu of the suspicious hard disk and start the “Error checking” in the “Tools” tab. In the following dialog box, you can specify whether you also want to check for bad sectors – this may take half an hour or more.
Alternatively, you can also use Checkdisk in the terminal, for example if you log on to the computer remotely or only have a repair console available. A typical call would be “chkdsk c: /f /r”, for example, to check drive “C:” for file system errors (/f) and bad sectors (/r) and have them repaired if necessary.