SSD S.M.A.R.T Attributes question

simmons33

Honorable
Nov 7, 2012
699
0
11,160
Ive been using a Kingston SSDNow V300 60GB SSD (With the 5.05 Firmware before Kingston screwed the pooch) since October 2013 and have a question regarding 3 S.M.A.R.T attributes Ive noticed both in Defraggler and Speccy recently..

Particularly the "195 On the fly ECC Uncorrectable Errors Count", "201 Uncorrectable Soft Read Error Rate", and "204 Soft ECC Correction Rate".

Im more or less paranoid these values indicate a potential failure. The drive still performs as great as it did the day I bought it, around 520R and 450W, Checkdisk says everything is fine... SMART itself says these values are "GOOD".

Anyone have any input? Ive scoured the web and really cant figure out what these 3 values mean.

ID | Attribute Name | Real Value | Current | Worst | Threshold | Raw Value

1 Read Error Rate 0 115 115 50 0x000005D9FA61
5 Retired Block Count 0 100 100 3 0x000000000000
9 Power-On Hours (POH) 658d 18h 82 82 0 0x000000003DC2
12 Power Cycle Count 989 100 100 0 0x0000000003DD
171 Program Fail Blocks Count 0 0 0 0 0x000000000000
172 Erase Fail Blocks Count 0 0 0 0 0x000000000000
174 Unexpected Power Loss 62 0 0 0 0x00000000003E
177 Wear Range Delta 11 0 0 0 0x00000000000B
181 Program Fails Count 0 0 0 0 0x000000000000
182 Erase Fails Count 0 0 0 0 0x000000000000
187 Reported Uncorrectable Errors 0 100 100 0 0x000000000000
189 High Fly Writes 81,608,704,029 29 66 0 0x00000042001D
194 Temperature 29 °C 29 66 0 0x00000042001D
195 On the fly ECC Uncorrectable Errors Count 98,171,489 120 120 0 0x000005D9FA61
196 Reallocation Event Count 0 100 100 3 0x000000000000
201 Uncorrectable Soft Read Error Rate 98,171,489 120 120 0 0x000005D9FA61
204 Soft ECC Correction Rate 98,171,489 120 120 0 0x000005D9FA61
230 Life Curve Status 100 100 100 0 0x000000000064
231 SSD Life Left 0 91 91 10 0x000000000000
233 Media Wearout Indicator 31,773 0 0 0 0x000000007C1D
234 Average Erases Count, Max Erases Count 15,938 0 0 0 0x000000003E42
241 Lifetime Writes From Host 15,938 0 0 0 0x000000003E42
242 Lifetime Reads From Host 16,383 0 0 0 0x000000003FFF
 
Solution
It's just like RAM, EEC is the error correction. These alone generally are cause for alarm, nor indication of a failing drive. The other values look fine to me.

My usual recommendation on SSD is just use it like any other hard drive, don't sweat read / writes. If you want to look at the SMART data every now and then fine. In reality most quality SSD drives will last longer than you will need them for. They are only getting cheaper and faster. By the time that drive fails, there will be something faster / larger out that you'll want to upgrade to anyways. As with anything though you should have backups. Any drive, even with a perfect SMART report, could fail at any random time without warning.
It's just like RAM, EEC is the error correction. These alone generally are cause for alarm, nor indication of a failing drive. The other values look fine to me.

My usual recommendation on SSD is just use it like any other hard drive, don't sweat read / writes. If you want to look at the SMART data every now and then fine. In reality most quality SSD drives will last longer than you will need them for. They are only getting cheaper and faster. By the time that drive fails, there will be something faster / larger out that you'll want to upgrade to anyways. As with anything though you should have backups. Any drive, even with a perfect SMART report, could fail at any random time without warning.
 
Solution
Managed to find a .pdf and post from Kingston that claims the best normalized value for all 3 of those attributes is 120. I would assume without doing the math that those values would decrease from 120 down to a worst of 38 if the drive was actually having errors.. and that the real value is simply the number of reads/writes since the drive has been powered on, and would be cleared once the drive is powered off.

http://media.kingston.com/support/downloads/MKP_306_SMART_attribute.pdf

Also found that Kingston actually created a new SSD management tool called Kingston SSD Manager (RIP SSD Toolbox) and after comparing the values in the PDF with those found in the S.M.A.R.T readout in SSD Manager along with the fancy colored bars, everything checks out as completely fine.

Im aware S.M.A.R.T may not issue a warning before a drive failure, but still nice to know.

Actually, the last 5 drives Ive had fail over the last few years threw up S.M.A.R.T warnings shorty before kicking the bucket...so I tend to check from time to time.
 
You have pretty bad luck, 5 failed drives? I dont want to jinx myself, but every drive Ive ever bought is still running. I even have one of the original 36 GB Raptor drives still running in my inlaws PC.

Got two Corsair SSDs in my desktop, Mushkin SSDS in both my HTPCs. Everything else is on WD Black drives.
 


3 were in a Dell XPS M1530, WD Scorpio Blacks, had one die in 2006, 2007 and 2008 the other 2 were drives that lasted 6 years of near 24/7 usage before going kaput

Probably should have mentioned the timeline =P