Raw Read Error Rate

[jdog2pt0]

As I am not sure what all this data means, an online SMART scan through speedfan gives me this.
Attribute Current Raw Overall
0 Raw Read Error Rate 163 175824 Watch
Warning: Raw Read Error Rate is below the average limits (173-200).
10 Spin Up Time 175 2216 Very good
10 Start/Stop Count 100 7 Very good
10 Reallocated Sector Count 200 0 Very good
10 Seek Error Rate 200 0 Very good
6 Power On Hours Count 98 1605 Good
10 Spin Retry Count 100 0 Very good
10 Calibration Retry Count 100 0 Very good
10 Power Cycle Count 100 690 Very good
10 Power Off Retract Count 200 661 Very good
10 Load Cycle Count 200 697 Very good
1 Reallocated Event Count 199 1 Normal
10 Current Pending Sector 200 0 Very good
10 Offline Uncorrectable Sector Count 200 0 Very good
10 Ultra DMA CRC Error Rate 200 2 Very good
10 Write Error Rate 200 0 Very good

I have a Western Digital 2500AAKS and twice I have run the Data Lifeguard Diagnostics and gotten code 0223 or 223 here http://support.wdc.com/techinfo/general/errorcodes.asp . The first time ran I the diagnostic because I lost windows. The second time, just the other day, I can close to losing windows but managed to save it. The raw read error rate doesn't seem to be well on the smart scan and I would like to know more about this.

 

[jdog2pt0]

Sorry about the smart thing, nobody will be able to understand that, here's a picture instead.

 

[sub mesa]

No immediate problems. Only 1 reallocated sector, just the ECC needs to fix the raw medium errors, which is normal given the data density. Harddrives don't really know what they are storing anymore, only after ECC has been applied is the data reliable enough not to have any bit errors. In other words, its normal that the raw read error count is quite high. All modern disks are vulnerable to this issue.

 

[jdog2pt0]

At what point will the ECC kick in though

 

[jdog2pt0]

Kick not quick sorry, apparently I'm not allowed to edit my own post.

 

[sub mesa]

All times, without ECC your harddrive doesn't know what its storing. No modern harddrive does, they have to guess using ECC that the final result is without error, but even this cannot be guaranteed. HDDs sometimes really can't tell if a bit had been 0 or 1, which is called a bit flip. It can happen over time, and depends alot on the data density which is getter bigger and bigger. To the point that it gets harder and harder to know exactly what data it stored.

This problem may even be bigger on 1.5TB-2.0TB disks, but its something all modern HDDs have to deal with. Without ECC they can't be sure of the actual contents of the data, and would make MANY mistakes to the point of the HDD being unusable. So yeah, its normal.. kind of. Its not a good thing though.