Hard drive wearout, SSD Media Wearout Indicator

[wmltho...@gmail.com]

Another topic that came up at last nights meeting was drive mean time between failures[1]. There is also a mean time to failure, MTTF. Sadly not many hard drives provide this information. I do not believe any spinning disk drives have such. Some solid state hard drives, SSDs, have a indicator for such,

233 Media_Wearout_Indicator

It will not tell you exactly when the drive will fail. but gives you some idea that the media is start to wear out.  It might be something to look for when purchasing SSDs. You can find out if your drive supports this by the following, where /dev/sda is the drive, it could be another, but NOT a partition, the entire drive.

smartctl -data -A /dev/sda

Thus far I have only seen Intel SSDs have a 233 Media_Wearout_Indicator ( tried Seagate and Samsung SSDs and they do not have such ), as follows

Intel SSD

ws wlt # smartctl -data -A /dev/sda

smartctl 6.3 2014-07-26 r3976 [x86_64-linux-3.18.5-gentoo] (local build)

Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===

SMART Attributes Data Structure revision number: 10

Vendor Specific SMART Attributes with Thresholds:

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  5 Reallocated_Sector_Ct   0x0032   100   100   000    Old_age   Always       -       0

  9 Power_On_Hours_and_Msec 0x0032   000   000   000    Old_age   Always       -       902124h+24m+41.070s

 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       595

181 Program_Fail_Cnt_Total  0x0032   000   000   000    Old_age   Always       -       0

182 Erase_Fail_Count_Total  0x0032   000   000   000    Old_age   Always       -       0

192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       594

225 Host_Writes_32MiB       0x0032   100   100   000    Old_age   Always       -       77658

232 Available_Reservd_Space 0x0033   100   100   010    Pre-fail  Always       -       0

233 Media_Wearout_Indicator 0x0032   100   100   000    Old_age   Always       -       0

241 Host_Writes_32MiB       0x0032   100   100   000    Old_age   Always       -       77658

242 Host_Reads_32MiB        0x0032   100   100   000    Old_age   Always       -       24506

249 NAND_Writes_1GiB        0x0013   100   100   000    Pre-fail  Always       -       2978

Samsung SSD

dev ~ # smartctl -data -A /dev/sda

smartctl 6.3 2014-07-26 r3976 [x86_64-linux-3.17.4-hardened-r1] (local build)

Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===

SMART Attributes Data Structure revision number: 1

Vendor Specific SMART Attributes with Thresholds:

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail  Always       -       0

  9 Power_On_Hours          0x0032   097   097   000    Old_age   Always       -       10585

 12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       134

177 Wear_Leveling_Count     0x0013   099   099   000    Pre-fail  Always       -       6

179 Used_Rsvd_Blk_Cnt_Tot   0x0013   100   100   010    Pre-fail  Always       -       0

181 Program_Fail_Cnt_Total  0x0032   100   100   010    Old_age   Always       -       0

182 Erase_Fail_Count_Total  0x0032   100   100   010    Old_age   Always       -       0

183 Runtime_Bad_Block       0x0013   100   100   010    Pre-fail  Always       -       0

187 Uncorrectable_Error_Cnt 0x0032   100   100   000    Old_age   Always       -       0

190 Airflow_Temperature_Cel 0x0032   073   065   000    Old_age   Always       -       27

195 ECC_Error_Rate          0x001a   200   200   000    Old_age   Always       -       0

199 CRC_Error_Count         0x003e   099   099   000    Old_age   Always       -       38

235 POR_Recovery_Count      0x0012   099   099   000    Old_age   Always       -       82

241 Total_LBAs_Written      0x0032   099   099   000    Old_age   Always       -       915541701

Seagate SSD

wlt ~ # smartctl -data -A /dev/sda

smartctl 6.3 2014-07-26 r3976 [x86_64-linux-3.18.5-gentoo] (local build)

Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===

SMART Attributes Data Structure revision number: 0

Vendor Specific SMART Attributes with Thresholds:

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  1 Raw_Read_Error_Rate     0x000e   166   166   006    Old_age   Always       -       21

  5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always       -       0

  9 Power_On_Hours          0x0032   092   092   000    Old_age   Always       -       7156

 12 Power_Cycle_Count       0x0032   100   100   020    Old_age   Always       -       786

171 Unknown_Attribute       0x0032   100   100   000    Old_age   Always       -       0

172 Unknown_Attribute       0x0032   100   100   000    Old_age   Always       -       0

181 Program_Fail_Cnt_Total  0x0032   100   100   000    Old_age   Always       -       0

182 Erase_Fail_Count_Total  0x0032   100   100   000    Old_age   Always       -       0

194 Temperature_Celsius     0x0022   038   000   000    Old_age   Always       -       38 (Min/Max 13/58)

201 Unknown_SSD_Attribute   0x0032   100   100   000    Old_age   Always       -       0

204 Soft_ECC_Correction     0x0032   100   100   000    Old_age   Always       -       0

231 Temperature_Celsius     0x0013   097   097   010    Pre-fail  Always       -       4

234 Unknown_Attribute       0x0032   100   100   000    Old_age   Always       -       33063

241 Total_LBAs_Written      0x0032   100   100   000    Old_age   Always       -       1792

242 Total_LBAs_Read         0x0032   100   100   000    Old_age   Always       -       1029

250 Read_Error_Retry_Rate   0x0032   100   100   000    Old_age   Always       -       1407

And for fun, here is a spinning disk drive, last one I have in use outside of gaming platforms (ps4/xbox1)

Seagate spinning disk

dev ~ # smartctl -data -A /dev/sdc

smartctl 6.3 2014-07-26 r3976 [x86_64-linux-3.17.4-hardened-r1] (local build)

Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===

SMART Attributes Data Structure revision number: 10

Vendor Specific SMART Attributes with Thresholds:

ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE

  1 Raw_Read_Error_Rate     0x002f   118   099   006    Pre-fail  Always       -       171239328

  3 Spin_Up_Time            0x0023   097   097   000    Pre-fail  Always       -       0

  4 Start_Stop_Count        0x0032   100   100   020    Old_age   Always       -       99

  5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always       -       0

  7 Seek_Error_Rate         0x002f   100   253   030    Pre-fail  Always       -       557759

  9 Power_On_Hours          0x0032   089   089   000    Old_age   Always       -       10376

 10 Spin_Retry_Count        0x0033   100   100   097    Pre-fail  Always       -       0

 12 Power_Cycle_Count       0x0032   100   100   020    Old_age   Always       -       96

180 Unknown_HDD_Attribute   0x002b   100   100   000    Pre-fail  Always       -       0

183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always       -       0

184 End-to-End_Error        0x0033   100   100   097    Pre-fail  Always       -       0

187 Reported_Uncorrect      0x0032   100   100   000    Old_age   Always       -       0

188 Command_Timeout         0x0032   100   099   000    Old_age   Always       -       1 1 1

189 High_Fly_Writes         0x003a   100   100   000    Old_age   Always       -       0

190 Airflow_Temperature_Cel 0x0022   070   063   045    Old_age   Always       -       30 (Min/Max 25/32)

191 G-Sense_Error_Rate      0x0032   100   100   000    Old_age   Always       -       0

192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       52

193 Load_Cycle_Count        0x0032   099   099   000    Old_age   Always       -       3787

194 Temperature_Celsius     0x0022   030   040   000    Old_age   Always       -       30 (0 19 0 0 0)

196 Reallocated_Event_Count 0x0032   100   100   036    Old_age   Always       -       0

197 Current_Pending_Sector  0x0032   100   100   000    Old_age   Always       -       0

198 Offline_Uncorrectable   0x0030   100   100   000    Old_age   Offline      -       0

199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       0

1. http://en.wikipedia.org/wiki/Mean_time_between_failures

 

[Tim Holloway]

On spinning hard disks, as on most hardware, MTTF has no bearing on a specific unit - it's the Arithmetic Mean of the lifespans of all drives in the sample. Estimated MTTF is computed by extrapolated test-to-destruction and/or stats from similar product lines.

I haven't found it that hard to get MTTF's, but if you can't, just figure that it's somewhat longer than the warranty period, since excepting maybe ViewSonic, most manufacturers don't like handing out free replacements for stuff that dies of old age. Most likely 110-150% of the warranty period.

Of course, Your Mileage May Vary. A certain run of WD disks several years back had an in-house MTTF of about 7 days and I still haven't forgiven the network administrator for not running backups on the LAN.

On the mousetech server farm, I mostly use Seagates, and figure that 2-3 drives will die over a year's time. Seagate used to offer exceptionally long warranties, though not so much these days.

If SSDs want to report an estimated MTTF for their individual units it's probably because it's easier to estimate, since the cells in the solid-state array wear out after a limited number of uses, unlike hard drives, which are more at the mercy of things like motor wear and drive electronics.

Hybrid SSDs might or might not wish to estimate remaining lifespan, since depending on the design, the unit might simply shift more and more storage to the spinning platters as the solid-state memory ceases to operate.

--
You received this message because you are subscribed to the Google Groups "JaxLUG" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jaxlug+un...@googlegroups.com.
To post to this group, send email to jax...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.