Smart attributes

[gtaillandier]

Hello

Sorry for my English but I'm French.
I just want some précisions about SMART attributes
I don't understant the different values even after reading help file.
Attribute 1 : Threshold 51, value 200, worst 200. Data 111.

Can someone help me describing these attributes ?

Thank you very much

[hdsentinel]

Hello,

Please check Support -> Knowledge Base -> S.M.A.R.T. function
(or directly at http://www.hdsentinel.com/smart/index.php

This generally describes how things work: what the different fields mean: the "Data" is a raw number (measured by a sensor / internal counter), the Value is calculated based on this and usually decreases. The Worst shows the lowest Value ever found and the Threshold is the limit specified by the manufacturer: the attribute (if it's critical) indicates error if the Value decreases below the Threshold.

There are numerous problems with this approach, so we would need to have a more precise, better method which can be used to verify the actual REAL status.
This is included in Hard Disk Sentinel - the background is described on http://www.hdsentinel.com/smart/index.php

[gtaillandier]

Hello

Thank you for the answer. Even after read http://www.hdsentinel.com/smart/index.php, I dont' understand the different numbers for SMART attributes.

attri 1 Raw Read Error Rate

Threshold 51
Value 200
Worst 200
Status OK
Data 137.

Can you explain me that ?

Thank you very much

[hdsentinel]

Generally, as described on http://www.hdsentinel.com/smart/index.php
the Threshold field is determined by the manufacturer for each attributes. These are fixed and never change.

The Value field has a starting "best" value, it is 200 in this case. This is a theoretical maximum and upon problems, this decreases.
As you can see, the Worst field is also 200 - indicating that the lowest (=worst) value is also 200, so this attribute was never "worse" than the current status.

Please note that these numbers are for this hard disk only - an other hard disk (especially from an other manufacturer) may work completely differently and may have completely different numbers in all fields.

The Data field is a "raw" number. It can be a counter or a number measured by a sensor, depending on the meaning of the attribute and its purpose (you can check this on the bottom, next to the graph. If you select different attributes, you may see Error-rate, statistical, event count and so).

In most cases, the Value is calculated somehow by the Data field. For example, if there would be high number of errors, the Value may drop below the Threshold - indicating that hard disk drive failure may be predicted.
The problem is that in most cases, this method is not sensitive enough - as the change of the Value may be too late - when hard disk problems (and partial / complete) data loss occured.

This is why Hard Disk Sentinel focuses mostly on the "raw" numbers and determines the health based on these. It checks the "Value" / "Threshold" fields as well - but to get more accurate status reporting, the "raw" numbers used.

However, I can confirm that not all numbers in the "raw" field are errors. As described in the Help -> Components of the application -> Main window -> Physical drives -> S.M.A.R.T. :

Many attributes (for example, power on time, drive power cycle count, etc.) are always increasing. This will make the trend of the graph for these attributes to increase also, but it is completely normal. For some (especially statistical) attributes, the number can be very high and then it can fall, it is normal also. The graph is important for the critical attributes (for example reallocated, bad sector count) where the analysis of the previous values can reveal problematic days or period in the lifetime of the disk.

Of course I can confirm that if there would be a real error with this (or other) attribute, Hard Disk Sentinel would surely
- report in the text description
- show the degradation health value
- it would be detected by Disk -> Short self test or Disk -> Extended self test (you may try these, just to confirm)
- the Disk -> Surface test -> Read test could not be completed without problems (you may try this, just to confirm)

[gtaillandier]

Hello

In my case, attribut 1 has :
Threshold 51
Value, Worst 200
Status OK
Data 137.

In your answer, you write that Threshold is also 200. I'm sorry, but I don't understand.

[hdsentinel]

Sorry, I meant that "Worst" is 200 - as described in the text, so I corrected that line in my previous post.
("As you can see, the Worst field is also 200 - indicating that the lowest (=worst) value is also 200, so this attribute was never "worse" than the current status.")