Samsung SpinPoint T166 SMART attribute 187

[Andre Majorel]

This new Samsung SpinPoint T166 (HD501L, firmware CR100-10)
behaves strangely. I have a program that repeatedly { fills a
block device with pseudo-random data, reads it back and checks
that it matches }. I started the program last night on the
Samsung and this morning the console was full of kernel messages
and the disk was going clock... clock... clock...

Killed the program and the noise stopped. Restarted the program
and no noise, no kernel messages, but the writing speed was 10
MB/s instead of about 70 MB/s. There is another instance of the
program running on another disk on the same SATA controller.
That one was limited to about 7 MB/s. Moved the Samsung to
another SATA controller returned the write speeds on both disks
to more normal figures. Still no noise or kernel messages.

According to smartctl 5.36, SMART attribute 187 on the Samsung
has changed significantly. Last night it was :

ID# ATTRIBUTE_NAME FLAG VAL WORST THR TYPE UPDATED WF RAW_VALUE
1 Raw_Read_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0
3 Spin_Up_Time 0x0007 100 100 015 Pre-fail Always - 7104
4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 3
5 Reallocated_Sector_Ct 0x0033 253 253 010 Pre-fail Always - 0
7 Seek_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0
8 Seek_Time_Performance 0x0025 253 253 015 Pre-fail Offline - 0
9 Power_On_Hours 0x0032 253 253 000 Old_age Always - 0
10 Spin_Retry_Count 0x0033 253 253 051 Pre-fail Always - 0
11 Calibration_Retry_Count 0x0012 253 253 000 Old_age Always - 0
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 3
187 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0
188 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0
190 Unknown_Attribute 0x0022 079 079 000 Old_age Always - 21
194 Temperature_Celsius 0x0022 175 175 000 Old_age Always - 21
195 Hardware_ECC_Recovered 0x001a 100 100 000 Old_age Always - 223453
196 Reallocated_Event_Count 0x0032 253 253 000 Old_age Always - 0
197 Current_Pending_Sector 0x0012 253 253 000 Old_age Always - 0
198 Offline_Uncorrectable 0x0030 253 253 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0
200 Multi_Zone_Error_Rate 0x000a 253 253 000 Old_age Always - 0
201 Soft_Read_Error_Rate 0x000a 253 253 000 Old_age Always - 0
202 TA_Increase_Count 0x0032 253 253 000 Old_age Always - 0

This morning :

ID# ATTRIBUTE_NAME FLAG VAL WORST THR TYPE UPDATED WF RAW_VALUE
1 Raw_Read_Error_Rate 0x000f 100 100 051 Pre-fail Always - 7
3 Spin_Up_Time 0x0007 100 100 015 Pre-fail Always - 7104
4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 3
5 Reallocated_Sector_Ct 0x0033 253 253 010 Pre-fail Always - 0
7 Seek_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0
8 Seek_Time_Performance 0x0025 253 253 015 Pre-fail Offline - 0
9 Power_On_Hours 0x0032 253 253 000 Old_age Always - 11
10 Spin_Retry_Count 0x0033 253 253 051 Pre-fail Always - 0
11 Calibration_Retry_Count 0x0012 253 253 000 Old_age Always - 0
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 3
187 Unknown_Attribute 0x0032 001 001 000 Old_age Always - 9210
188 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0
190 Unknown_Attribute 0x0022 063 058 000 Old_age Always - 37
194 Temperature_Celsius 0x0022 127 112 000 Old_age Always - 37
195 Hardware_ECC_Recovered 0x001a 100 100 000 Old_age Always - 283627933
196 Reallocated_Event_Count 0x0032 253 253 000 Old_age Always - 0
197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 1
198 Offline_Uncorrectable 0x0030 253 253 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 1
200 Multi_Zone_Error_Rate 0x000a 100 100 000 Old_age Always - 0
201 Soft_Read_Error_Rate 0x000a 100 100 000 Old_age Always - 1
202 TA_Increase_Count 0x0032 253 253 000 Old_age Always - 0

 

[Arno Wagner]

This new Samsung SpinPoint T166 (HD501L, firmware CR100-10)
behaves strangely. I have a program that repeatedly { fills a
block device with pseudo-random data, reads it back and checks
that it matches }. I started the program last night on the
Samsung and this morning the console was full of kernel messages
and the disk was going clock... clock... clock...

Killed the program and the noise stopped. Restarted the program
and no noise, no kernel messages, but the writing speed was 10
MB/s instead of about 70 MB/s. There is another instance of the
program running on another disk on the same SATA controller.
That one was limited to about 7 MB/s. Moved the Samsung to
another SATA controller returned the write speeds on both disks
to more normal figures. Still no noise or kernel messages.

According to smartctl 5.36, SMART attribute 187 on the Samsung
has changed significantly. Last night it was :

Hmm. You have one pending sector (read: likely unreadable, but the
disk will continue trying) and a CRC inteface error. The CRC
error can well be the reason for the slow speed: The kernel reduces
the interface speed when errors occure on the cable. IT does thif
for the whole channel or controller.

As to the pending sector, that can happen with a new disk. It will be
the reason for the retry-sound. Disks are not completely
surface-checked with more sensitive equipment that the disk has today.

My advice you be to track down the CRC error first (bad cabeling?)
and then run your tool again. Alternatively run it on the other
controller. If nothing shows up, then you probably need not be
concerned.

Arno

ID# ATTRIBUTE_NAME FLAG VAL WORST THR TYPE UPDATED WF RAW_VALUE
1 Raw_Read_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0
3 Spin_Up_Time 0x0007 100 100 015 Pre-fail Always - 7104
4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 3
5 Reallocated_Sector_Ct 0x0033 253 253 010 Pre-fail Always - 0
7 Seek_Error_Rate 0x000f 253 253 051 Pre-fail Always - 0
8 Seek_Time_Performance 0x0025 253 253 015 Pre-fail Offline - 0
9 Power_On_Hours 0x0032 253 253 000 Old_age Always - 0
10 Spin_Retry_Count 0x0033 253 253 051 Pre-fail Always - 0
11 Calibration_Retry_Count 0x0012 253 253 000 Old_age Always - 0
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 3
187 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0
188 Unknown_Attribute 0x0032 253 253 000 Old_age Always - 0
190 Unknown_Attribute 0x0022 079 079 000 Old_age Always - 21
194 Temperature_Celsius 0x0022 175 175 000 Old_age Always - 21
195 Hardware_ECC_Recovered 0x001a 100 100 000 Old_age Always - 223453
196 Reallocated_Event_Count 0x0032 253 253 000 Old_age Always - 0
197 Current_Pending_Sector 0x0012 253 253 000 Old_age Always - 0
198 Offline_Uncorrectable 0x0030 253 253 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0
200 Multi_Zone_Error_Rate 0x000a 253 253 000 Old_age Always - 0
201 Soft_Read_Error_Rate 0x000a 253 253 000 Old_age Always - 0
202 TA_Increase_Count 0x0032 253 253 000 Old_age Always - 0

This morning :