I'm trying to run an extended test on a 2TB external disk connected to my laptop via a 7 port USB hub. The test doesn't complete because it's "interrupted by host with hardware/software reset ". I unchecked the power saving setting on the USB hub, but the error still occurs.
The external device information is:
Generally the short self test, extended self tests are hardware self tests, running "inside" the hard disk drives, completely out of control of any software. We can start the test and read the progress and results.
The message you wrote means that during the self test, the disk controller (or the USB adapter/converter/docking station) may enter the drive to standby mode because it does not "know" that disk test is running. It results that the disk test is aborted and a bus reset happens.
In some cases also the internal Advanced Power Management function of the hard disk can activate.
Usually the USB _HUB_ itself does not cause troubles, does not affect the operation (maybe, but there is small chance for that).
This does not mean any problem with the hard disk (or software).
To improve the situation, you may try to
- disable hard disk spin down timeout and power management in Windows
- if the hard disk supports Advanced Power Management (so the option Disk menu -> Advanced Power Management is not greyed in Hard Disk Sentinel) then select that function and drag the slider to right to set highest performance without standby
These may help to keep the drive spinning and active during the self test.
In worst case, you may try to connect the hard disk drive to different controller (eg. connect to internal SATA port) or use Disk menu -> Surface test -> Read test to verify the disk drive in this case (as suggested by the Help in case the hardware self tests may not working, as this can happen).
I'm running some software (Prevent Disk Sleep) which is supposed to keep the disk(s) active, but it doesn't seem to work. It activates every 500 seconds. Maybe I should shorten this period.
Unfortunately the disk in question doesn't have Advanced Power Management as the menu entry is greyed out, so I'll go to the worst case solution which is to do a Read Surface test.
The reason for all this trouble is that I recently started having BSOD after the laptop goes into sleep mode, but it does not not happen every time. The Windows error stop code is WIN32K POWER WATCHDOG TIMEOUT. I haven't found any useful information on the Internet about this stop code. It has been suggested that a it may be due to an external device hence I'm looking into this hard disk as a start, because I noticed that the backups that I save to it are relatively slow. Also the monitor may be the cause of the BSOD problem, but I haven't yet looked into it.
There is no reason to use any other tools. Hard Disk Sentinel of course automatically has this feature for many years.
Please open Configuration -> Disk Control and there is an option "Do not allow spin down of disks".
Just enter the disk number (as you can see in the upper left area of the main window) or specify a * (star) character to keep all drives spinning.
This periodically sends a command to the drive(s), keeping them awake (and preventing load/unload cycles, as this is also a frequent problem).
I suspect this BSOD will not really related to the disk drive. Maybe a problematic driver / update is in the background.
But yes, the surface test -> Read test can be better: as the Help describes, it can show both performance degradations (slower accessible sectors) and other issues which may be not reported by a simple hardware self test. Also it can confirm if the drive is really working as should, so the issue will be surely related to something else.
Thank you for pointing out the feature to not allow disks to spin down! I'm very happy to learn that HD Sentinel already has this.
I forgot to mention that the problem arose a few weeks ago and what I remember is that it was at about the time when there was an update to the graphics card driver of my laptop. I cannot say for sure if that's the cause of the Windows error, but my suspicions are focused on that event as I don't recall anything else significant at the time.
I'll do a Surface Read test as soon as possible. The last one that I did for the particular disk may be two months ago showed 100% health and performance.
