Bug or Issue?

How, what, where and why - when using the software.
PeteB
Posts: 6
Joined: 2014.03.13. 01:48

Bug or Issue?

Post by PeteB »

Just received an alert eMail for a hard disk in my system indicating predicted failure.

It was for attribute 197 - Current Pending Sector Count. The eMail had the following values: Threshold (200), Value (100), Worst (100), Data (000000000000). My understanding is that as the value is less than the threshold this is the reason for the alert.

In checking HDSentinel on the computer, the values I see are: Threshold (0), Value (100), Worst (100), Data (000000000000). The graph of pending sectors also shows 0 pending sectors for each day from the day the hard disk was installed. So HDSentinel main window does not agree with the eMail alert.

I also had an eMail alert for attribute 7 a few days earlier. The eMail had the following values: Threshold (248), Value (100), Worst (100), Data (000000000000).

However, in checking HDSentinel on the computer, the smart attributes for seek error rate are: Threshold (67), Value (100), Worst (100), Data (000000000000) and the attribute graph shows the seek error rate as continuously at 0.

I have checked the alert window for alerts and found the following:

23/12/2015 12:00:14 AM,Failure Predicted !!! Attribute : Current Pending Sector Count
20/12/2015 12:00:23 AM,"Health: 0 %, Limit: 25 % : Red health level reached."
20/12/2015 12:00:13 AM,Failure Predicted !!! Attribute : Seek Error Rate

I'm wondering what is going on here. This is only happening on the one hard disk. Do I have any reasons for concern?

HDSentinel status windows show the drive as perfect with a 100% health value. The drive details are as follows:
Hard Disk Model ID: HGST HDN724030ALE640
Firmware Revision: MJ8OA5E0
User avatar
hdsentinel
Site Admin
Posts: 3128
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: Bug or Issue?

Post by hdsentinel »

Thanks for your message.

Yes, it's definitely an issue with the storage system, not really a bug of Hard Disk Sentinel.

The problem is that from time-to-time the software may got improper response when the hard disk status accessed and detected.
It is not frequent, but it may happen (for example once per every 1000 detections or so) - and if this happens, it triggers a false alarm.

In most cases, this is not really caused by the hard disk itself, but more related to the disk controller (motherboard) where the hard disk connected, more precisely its current driver, which may sometimes provide such invalid status information. According the details you wrote, the error-level threshold numbers seem inaccurate, but this is enough to trigger a false alarm when the thresholds seem reached.

Generally Hard Disk Sentinel has numerous filters against such situations and constantly investigating different controllers, drivers and hard disk combinations to prevent such. The Support -> Driver Zone page designed to offer the best working drivers for different controllers / chipsets where there should be problems. Note that the "best" and "latest" are usually not the same...

I'm not sure about the current Hard Disk Sentinel version (displayed on the top), the disk controller and its driver (displayed on the Information page when the hard disk selected).
If you use older Hard Disk Sentinel, I'd recommend to update to the latest possible version and you may also consider a different driver for the controller - if possible.

Also please use Report menu -> Send test report to developer option any time, as then it is possible to check the actual situation, the "raw" information provided by the disk controller (and its driver) and improve the filter to prevent such false alarms on this system. Excuse me for that, I hope by the developer report things can be improved quickly.
Post Reply