weird's in SMART explanations

[Ernst@at]

0xC6/198 Off-line uncorrectable sector count:
"Count of uncorrectable errors when reading/writing"
Under this code afaik the read errors detected while surface scan in idle mode are accounted, alike the 0xC5/197 in active read mode.

0xc8/200 Write Error Rate
"Errors occured while writing raw data from a disk, ..." - afaik writing is done to a disk

PS: The Help file does not explain the (later added ?) function of "Configuration: Disk control" - "Do not allow spin down ..."
What type of keep awake is there implemented? Only Control Commands? In which interval? Some strange firmware parks the head after a few seconds of inactivity - can this be prohibited with this option?
Or does it influence APM settings?

[hdsentinel]

Thanks for your attention and message!

Yes, I agree that in theory the 0xC6/198 Off-line uncorrectable sector count should show errors detected in "offline disk test" (for example the hardware self test invoked by Disk -> Extended self test). However, according the experiences and reports, many drives use this attribute to reflect the count of reallocated sectors (even found during normal operations) instead of using attribute #5 for this purpose.

0xC5/197 Current pending sectors count shows the number of sectors which have CRC damaged and the drive would need to re-check the status (during write command(s) targeted to the sector) and reallocate if required or restore the stable status of the sector.


0xc8/200 Write Error Rate
Absolutely correct: writing TO a disk. Sorry for the mistake, it will be corrected of course, thanks for the tip!


PS.: yes, the "Do not allow spin down ..." function is added after the help last revised, this is why it is not yet included.
This option is used to periodically send control command to prevent both spin down and parking of the heads.
Hard Disk Sentinel checks both the drive type and connection type (because some USB adapters / enclosures have their own timer to force spin down / park after some seconds of inactivity) and automatically send the commands on required intervals. This can be once per every 4 seconds to once per every 30 seconds, depending on the actual situation.
Also yes, I confirm that the APM setting (can be adjusted from the Disk menu - if not blocked by the hard disk controller driver) may be useful to prevent parking and spin down.

[celavey]

So this means that the faster the connection the better.

[hdsentinel]

Excuse me, I do not really understand what do you mean here with this ....

So this means that the faster the connection the better.