Hi,
Thanks for your message and the information.
May I ask why you selected the "Bug reports" section? To be honest, I did not see any problems (bugs) in Hard Disk Sentinel, so I do not really understand why this post is here.
> I am user of Hard Disk sentinel and I'm wondering how correct the info provided by the hard drives really are.
Hard Disk Sentinel is designed to exactly to display the REAL, correct status information, exactly to help us to detect and display any (even minor) issues.
Not only critical ones, but also minor problems which may simply reduce performance and/or can indirectly cause data corruption / data loss (eg. high temperatures, end-of-lifetime and so).
And (apart from the current error counters and the current health displayed) Hard Disk Sentinel is designed to show the possible CHANGE in the current status - which can also help users to recognise that something happened which may require attention (and at least backup of important data).
> WDC's LifeGuard diagnostics
This is completely different - and designed for completely different purpose, exactly the opposite: to hide, minimise problems.
As you can check on the list of self-monitoring attributes, the list is not as detailed as in Hard Disk Sentinel: the actual "raw" data values (the real error counters, status information) is completely missing.
Manufacturer specific tools designed to minimise service time / maintenance costs: so they simply check if the amount errors reached the threshold for critical attributes - or not.
They provide no information if the error counter(s) increased, decreased or so - even if these are important to determine the actual status of the drive ( as described at
www.hdsentinel.com/smart )
> says that only a few smart info attributes are under warranty, namely Raw Read Error Rate, Spin Up Time and Rallocated Sector Count.
Yes. If you see these attributes, you may notice that only these has non-zero values in the Threshold column: and if the Value column (calculated based on the actual real errors) drop below these, the hard disk is considered to be fail - and warranty replacement is possible.
With other words, only these attributes can even trigger S.M.A.R.T. failure and other attributes (even if they may provide valuable information about the current status) are completely ignored by this tool - but not by Hard Disk Sentinel.
> And the value in "Value" is loosely calculated from the "Data" value. So even though the "Data" value in my case
> is 2865 the "Value" value is still 200 and this is correct.
Yes.
As
www.hdsentinel.com/smart page reflect, this method (checking thresholds-values only) is poblematic as it does not really reflect the actual status.
This is why it is important to verify the raw "Data" - and check if it may report problem or not.
Not only for the mentioned attributes - but also for ones which may be ignored by other tools but can reflect problems.
> But more surprisingly the number of Off-Line Uncorrectable Sectors had been reduced from 81 to only 3 -
> also the health of the drive was 95%. What kind of sectors are these "Off-Line Uncorrectable Sectors"?
Generally this counter should reflect the amount of bad (reallocated) sectors.
These sectors yes, as we'd expect, represent the amount of sectors which are no longer used - which are replaced by the spare area.
On most drives, this has similar meaning as the 5 Reallocated sectors count attribute - just (how interesting...) has 0 threshold, so the hard disk should never trigger S.M.A.R.T. failure based on this attribute and you can never ask for warranty replacement (sorry for that).
Please check:
http://www.hdsentinel.com/faq.php#health
which describes exactly the situation: when some bad sectors found, warrantly replacement may be not possible.
However, on many hard disks, this counter has a different meaning: shows the currently STILL used pending sectors.
According the image, I see that the other attribute "current pending sectors" shows similar value.
These pending sectors (weak sectors) are even worse than the above mentioned bad sectors as these are still used by the hard disk: data stored in these sectors may become corrupted / lost.
Fortunately, these weak sectors are usually not critical and in many cases they are caused by the current operating environment, not really by the hard disk itself. The most common causes of weak sectors are:
- drive failure (error with internal memory, problem with drive head or surface)
- power loss (the write operation was not finished because of power loss)
- power failure (weak power supply or not stable power line)
- data cable failure or improper connection
- system memory or motherboard problem
- overclocking
Because weak sectors may not indicate real failure of the drive, they are not counted as seriously as real bad sectors or other attributes / issues.
How interesting, completely ignored by the manufacturer specific tool - but counted and reported by Hard Disk Sentinel.
Yes, because the manufacturer specific tool focuses on the hard disk - while Hard Disk Sentinel focuses on your data.
To diagnose and repair such weak sectors, I always recommend to use Disk menu Surface test option in Hard Disk Sentinel.
The READ test may reveal the proper location of the sector and the WRITE test can be used to fix (repair) this problem. The best way is the "Reinitialise disk surface" test which fixes all such problems. This requires a previous backup of the disk as the contents are completely erased during this procedure.
In some cases, the format can be also helpful - but not as effective and some weak sectors may remain (as you could see).
Ideally, after testing, weak sectors completely eliminated and the status of the hard disk improves - even back to 100%.
Just it is important to know that if the original cause (cable, connection, etc.) is not fixed, then you may expect weak sectors again and again.
Because of the above, when weak sectors (pending sectors) detected, I NEVER recommend to send the drive to warranty.
Service may
1) perform a complete overwrite to "fix" the errors and then give back the drive as repaired.
This may take long time and they may even ask $$$ for repairing - and I'm not sure you'd feel safer if the drive is "repaired"
2) may give back an other hard disk, maybe with even worse health / condition
So instead, I always recommend to know more about your drive, verify the operating conditions, improve in all possible ways and perform the fix with Hard Disk Sentinel.
This is exactly described in the Support -> Frequently Asked Questions -> What is a weak sector? How to repair weak sectors?
section:
http://www.hdsentinel.com/hard_disk_cas ... ectors.php
> When a bad sector is remapped by the hard drive, the bad sector is hidden for software like Hard Disk Sentinel.
Absolutely NOT.
The original bad sector can't be read/written by any software (even by Hard Disk Sentinel) - but its count (and change in its count) is detected and displayed by Hard Disk Sentinel: both in the text description, the changing health % and on the Log page.
In contrast, in other tools, yes, it (and its count) may remain completely hidden - so without Hard Disk Sentinel you'd never know if your drive has one or more bad sectors.
> Is the remapping also hidden? I would imagine a remapping should be logged in SMART.
No, it is not hidden: the appropriate S.M.A.R.T. attributes reflect that.
Just different drives may report them differently - and this is why it is important to have a proper tool which can identify and detect, report the problems - and also allows diagnostics to reveal and fix the problems.
Just the manufacturer-specific tool hides those values.
> At the moment i think SMART and manufacturers implementation of SMART lacks the transparency that we as end users need.
This is exactly the purpose of Hard Disk Sentinel.
And this is exactly described at
www.hdsentinel.com/smart
even if manufacturers "play" with Thresholds/Values and use the attributes differently, we can find the way to detect and display the real error counters and determine the status of the drive based on those counters.
It is harder and harder to properly identify the meaning of S.M.A.R.T. attributes from time to time with the different hard disk models (and their different firmware versions which may also affect which attribute should reflect what kind of errors / problems).
And we did not mentioned OEM-specific hard disks, SSDs, SSHDs - where these are even more complicated.
Even if manufacturer-specific tools and other S.M.A.R.T. implementations (eg. the BIOS S.M.A.R.T. check) uses wrong methods (which are described on
www.hdsentinel.com/smart ), Hard Disk Sentinel can help - as it is designed to determine how the attributes used and reflect the overall status quickly and easily in the health % value and the text description - to be notified about any issues and allow time to perform backup and fix in all possible ways.
