I have several external USB drives on a mostly unattended Windows server.
I discovered one drive was non-responsive - attempting to access it caused Windows Explorer to go non-responsive.
Wondering what was going on with the drive, I tried to open HDSentinel Pro 5.70 to check it.
I was surprised to find it was also unresponsive.
Since I'm trusting HDSentinel to monitor my drives and alert me of errors and failures, ideally HDSentinel would be able to "survive" any HD error so it could alert the situation.
The server logs show a series of "The driver detected a controller error on \Device\Harddisk8\DR20"
I powered down the drive and rebooted the server, and now everything seems fine. HDSentinel reports the drive is "PERFECT", and there is nothing in the log, and nothing unusual in SMART.
Is there any setting or configuration option that would help keep HDSentinel functional regardless of the nature of the disk error (assuming the OS is still running, which it was)
USB Drive issue makes HDSentinel unresponsive
-
hdsentinel
- Site Admin
- Posts: 3128
- Joined: 2008.07.27. 17:00
- Location: Hungary
- Contact:
Re: USB Drive issue makes HDSentinel unresponsive
Exactly as you noticed "attempting to access it caused Windows Explorer to go non-responsive. "
This is true for anything else: any attempt to access the hard disk causes that the software/tool which tries to access it never got response from the disk drive, so the software trying to access seems "unresponsive" (frozen).
I'm afraid this is completely normal and there is no way to avoid it, as any software attempting to access (for example to detect its presence at all, status information and/or logical drive, files/folders etc.) needs to wait for the response of the device itself.
If you try to open Windows Explorer, Windows Disk Management or Hard Disk Sentinel - the situation is same: the software is waiting for the response and it can't be avoided.
The "freezing" itself is a clear indication that something is wrong/not stable in the storage system, as one or more device(s) can cause generic system instability too. Similarly that (for example) a memory test tool may freeze/hang completely when a failed memory module installed.
> I powered down the drive and rebooted the server, and now everything seems fine.
> HDSentinel reports the drive is "PERFECT", and there is nothing in the log, and nothing unusual in SMART.
This situation is very-very common when a 2.5" USB-powered hard disk is (or previously was) connected to USB 2.0 slot.
Then the disk drive may hard to spin up and start and may hardly respond or not respond at all due to insufficient power.
In this case, the hard disk can block everything (as you can see). Ideally it can start up, so a power cycle (or just disconnection/reconnection) may help and then everything seems working perfectly - but there is good chance that the hard disk drive will fail completely "without a notice" due to the insufficient power it receives.
Not sure if this is the case (if the hard disk was a 2.5" USB-powered hard disk drive). But if not, I'd also check the power supply/cables/connections.
Generally the page
https://www.hdsentinel.com/hard_disk_ca ... _retry.php
describes the situation with more details.
> Is there any setting or configuration option that would help keep HDSentinel functional regardless of the nature of the disk error (assuming the OS is still running, which it was)
I'm afraid no - because what you see is completely "normal" in such/similar case: Hard Disk Sentinel (when tries to access the disk drive to detect at least basic parameters) needs to wait for the response of the drive - which may never happen while it is battling for its life (to have proper spin up and begin normal operation first).
This is true for anything else: any attempt to access the hard disk causes that the software/tool which tries to access it never got response from the disk drive, so the software trying to access seems "unresponsive" (frozen).
I'm afraid this is completely normal and there is no way to avoid it, as any software attempting to access (for example to detect its presence at all, status information and/or logical drive, files/folders etc.) needs to wait for the response of the device itself.
If you try to open Windows Explorer, Windows Disk Management or Hard Disk Sentinel - the situation is same: the software is waiting for the response and it can't be avoided.
The "freezing" itself is a clear indication that something is wrong/not stable in the storage system, as one or more device(s) can cause generic system instability too. Similarly that (for example) a memory test tool may freeze/hang completely when a failed memory module installed.
> I powered down the drive and rebooted the server, and now everything seems fine.
> HDSentinel reports the drive is "PERFECT", and there is nothing in the log, and nothing unusual in SMART.
This situation is very-very common when a 2.5" USB-powered hard disk is (or previously was) connected to USB 2.0 slot.
Then the disk drive may hard to spin up and start and may hardly respond or not respond at all due to insufficient power.
In this case, the hard disk can block everything (as you can see). Ideally it can start up, so a power cycle (or just disconnection/reconnection) may help and then everything seems working perfectly - but there is good chance that the hard disk drive will fail completely "without a notice" due to the insufficient power it receives.
Not sure if this is the case (if the hard disk was a 2.5" USB-powered hard disk drive). But if not, I'd also check the power supply/cables/connections.
Generally the page
https://www.hdsentinel.com/hard_disk_ca ... _retry.php
describes the situation with more details.
> Is there any setting or configuration option that would help keep HDSentinel functional regardless of the nature of the disk error (assuming the OS is still running, which it was)
I'm afraid no - because what you see is completely "normal" in such/similar case: Hard Disk Sentinel (when tries to access the disk drive to detect at least basic parameters) needs to wait for the response of the drive - which may never happen while it is battling for its life (to have proper spin up and begin normal operation first).
Re: USB Drive issue makes HDSentinel unresponsive
Thanks for the insights. For reference, the drive is a 3.5" Hitachi 6T in a Pluggable USB3 dock (with a dedicated 12V "wall wart" switching power supply) connected to a a USB3 port.
I'm a basic level software developer. Clearly much of the blame falls on Windows - it seems like a bad design for a call to the OS to never return (even a failure) for hours or days. But it shouldn't have to be fatal for an application if the application is written with this issue in mind. If I were to "find a way to send an alert when a specific device/service/API is unavailable and hung", I might approach the problem by spawning a new process or thread with a semaphore and callback. If the new thread doesn't callback to report status (success/fail) in a reasonable amount of time, I'd send an alert using the still-functional capabilities of the OS, such as the IP stack for sending an email.
I'm a basic level software developer. Clearly much of the blame falls on Windows - it seems like a bad design for a call to the OS to never return (even a failure) for hours or days. But it shouldn't have to be fatal for an application if the application is written with this issue in mind. If I were to "find a way to send an alert when a specific device/service/API is unavailable and hung", I might approach the problem by spawning a new process or thread with a semaphore and callback. If the new thread doesn't callback to report status (success/fail) in a reasonable amount of time, I'd send an alert using the still-functional capabilities of the OS, such as the IP stack for sending an email.