Earlier this week we talked about hard drives and other mass storage devices in our Computer Buyer’s Guide. Today we discuss how to identify a failing hard drive or SSD by reading SMART attributes. The Self-Monitoring, Analysis and Reporting Technology monitors the health status of hard drives and SSDs and gives an indication that a device may be on the verge of failing. It’s not 100% accurate and does not tell you when a hard drive will fail, but if you’re considerate enough, it can help you decide when it’s time to repalce your hard drive. Note that some attributes are vendor-specific and may not be available on all drives. In this list, attributes marked in italics are unavailable on SSDs.
Depending on your operating system, how you access SMART readings is going to be different. A program available for all three major operating system is GSmartControl.
|Reallocated sector count||
Represents the number of sectors that couldn’t be written to and were marked as damaged by the SMART assessment. When this happens, the hard drive writes data to a sector in a special segment of the drive called a reserve pool.
|Spin retry count||The number of attempts to take the hard drive platters to full working speed. A high value may indicate issues with the drive’s mechanical system.|
|Reported uncorrectable errors||The number of errors that couldn’t be recovered by the drive’s error correction systems. A high value is an indication of a failing drive.|
|Unexpected power loss count||This value indicates how many times the drive’s arm couldn’t be properly repositioned because the computer was shut down in an unexpected manner, such as holding down the power button or because of a power outage. While not a failure indicator per se, high values of this attribute may lead to higher reallocated sector counts and spin retries. Some manufacturers may use other names, such as Power-off Retract Count, Emergency Retract Cycle Count, or Unsafe Shutdown Count.|
|Command timeout||How many times the execution of a command failed because the drive didn’t respond quickly enough. If the number is greater than zero, it indicates an issue with the power supply or the SATA cable.|
|Reallocation event count||The number of transfers of data from damaged sectors to the reserve pool. Both successful and unsuccessful events are counted.|
|Current pending sector count||If a value can’t be read, the hard drive “remembers” to look it up later to see if it can successfully read on successive attempt.|
|Uncorrectable sector count||The total number of sectors that couldn’t be corrected and, as such, were remapped to the reserve pool. A high value indicates issues with the disk surface or the motor.|