Is my Seagate about to fail?

[rry]

Seagate Model 7200.12, ST3500418AS

This drive is being used as the primary datastore and operating system volume on an ESXI 5.0.0 host, so quite a critical disk. (and yes, I know there are better enterprise grade drives to use) Recently I perceived some degradation in performance in one of my VM's, but there were no error messages to suggest anything untoward. So collected some SMART data from the disk drives, and whilst there are no errors reported from the short and extended testing, there are some very high SMART attribute values that may be on concern on this seagate. Only problem is that I understand that many of the raw values on Seagate drives can be ignored.

Whilst the drive continues to function at present, can the guru's out there advise whether the SMART attributes suggest failure is imminent and that I should replace the drive. Despite some research, I am not much the wiser.

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 0x000f 120 099 006 Pre-fail Always - 243450007
3 Spin_Up_Time 0x0003 097 097 000 Pre-fail Always - 0
4 Start_Stop_Count 0x0032 092 092 020 Old_age Always - 8778
5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always - 0
7 Seek_Error_Rate 0x000f 045 042 030 Pre-fail Always - 67629038668649
9 Power_On_Hours 0x0032 039 039 000 Old_age Always - 53691
10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 0
12 Power_Cycle_Count 0x0032 099 099 020 Old_age Always - 1634
183 Runtime_Bad_Block 0x0032 100 100 000 Old_age Always - 0
184 End-to-End_Error 0x0032 100 100 099 Old_age Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
188 Command_Timeout 0x0032 100 081 000 Old_age Always - 4295032991
189 High_Fly_Writes 0x003a 099 099 000 Old_age Always - 1
190 Airflow_Temperature_Cel 0x0022 065 049 045 Old_age Always - 35 (Min/Max 34/35)
194 Temperature_Celsius 0x0022 035 051 000 Old_age Always - 35 (0 24 0 0 0)
195 Hardware_ECC_Recovered 0x001a 038 017 000 Old_age Always - 243450007
197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0
198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0
240 Head_Flying_Hours 0x0000 100 253 000 Old_age Offline - 153768419171982
241 Total_LBAs_Written 0x0000 100 253 000 Old_age Offline - 967196094
242 Total_LBAs_Read 0x0000 100 253 000 Old_age Offline - 2324239304 (Note the 245% increase in this value for only an extra 6% in hours online)



for reference, here's some data from 3000 hours earlier
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
1 Raw_Read_Error_Rate 0x000f 117 099 006 Pre-fail Always - 139040010
3 Spin_Up_Time 0x0003 097 097 000 Pre-fail Always - 0
4 Start_Stop_Count 0x0032 092 092 020 Old_age Always - 8764
5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always - 0
7 Seek_Error_Rate 0x000f 043 042 030 Pre-fail Always - 63351144857163
9 Power_On_Hours 0x0032 043 043 000 Old_age Always - 50789
10 Spin_Retry_Count 0x0013 100 100 097 Pre-fail Always - 0
12 Power_Cycle_Count 0x0032 099 099 020 Old_age Always - 1627
183 Runtime_Bad_Block 0x0032 100 100 000 Old_age Always - 0
184 End-to-End_Error 0x0032 100 100 099 Old_age Always - 0
187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0
188 Command_Timeout 0x0032 100 081 000 Old_age Always - 4295032991
189 High_Fly_Writes 0x003a 099 099 000 Old_age Always - 1
190 Airflow_Temperature_Cel 0x0022 062 049 045 Old_age Always - 38 (Min/Max 30/38)
194 Temperature_Celsius 0x0022 038 051 000 Old_age Always - 38 (0 24 0 0 0)
195 Hardware_ECC_Recovered 0x001a 032 017 000 Old_age Always - 139040010
197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0
198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0
199 UDMA_CRC_Error_Count 0x003e 200 200 000 Old_age Always - 0
240 Head_Flying_Hours 0x0000 100 253 000 Old_age Offline - 184765198144302
241 Total_LBAs_Written 0x0000 100 253 000 Old_age Offline - 2343715752
242 Total_LBAs_Read 0x0000 100 253 000 Old_age Offline - 948144170

[data-medics]

I could be wrong here, but it looks like the values in lines 241 and 242 just switched places now.

Probably is just a quirk in reading the S.M.A.R.T. values. Unfortunately, manufacturers have never fully standardized S.M.A.R.T. values, so software is never perfect at reading and interpreting the values.

Given that it's a Seagate though, I'd say failure is imminent despite what SMART says

[fzabkar]

I could be wrong here, but it looks like the values in lines 241 and 242 just switched places now.

If that's true, then after 3000 hours the value would have decreased from 2343715752 to 2324239304. Clearly that's not possible, so ISTM that this is merely a coincidence.

[fzabkar]

IMHO the drive is in excellent condition given that it has racked up 50000 hours.

The normalised value of the Raw_Read_Error_Rate attribute is 120, which is the maximum possible score. AIUI, the way that this attribute is calculated is that the drive performs 250 million reads and records the number of read errors during this time. The raw value is the read count while the normalised value reflects the error rate. After each block of 250 million reads the raw value rolls over to 0 and the drive starts counting again. The raw value of the Hardware_ECC_Recovered attribute follows a similar pattern.

The Seek_Error_Rate attribute counts the number of seeks and seek errors during the drive's lifetime. The raw value of the attribute reports the lifetime seek count in the lower 32 bits and the lifetime seek error count in the uppermost bits.

Taking your example ...

67629038668649 = 0x3D82 1CD38B69

So the drive has recorded 15746 (= 0x3D82) seek errors in 483625833 (= 0x1CD38B69) seeks.

The normalised value is calculated as follows:

-10 log (0x3D82 / 0x1CD38B69) = 44.9

http://www.google.com/search?q=-10+log+(0x3D82+/+0x1CD38B69)

The Command_Timeout attribute is also best viewed in hexadecimal.

4295032991 = 0x0001 0001 009F

This suggests to me that the drive has recorded 159 (= 0x9F) timeouts. This number has not changed in 3000 hours, so perhaps the drive has stabilised since those errors were recorded.

ICBW, but the Total_LBAs_Written and Total_LBAs_Read attributes appear to be 32-bit values which roll over to 0 when the drive has read/written 4 billion (= 2^32) sectors. This corresponds to 2TB.

BTW, is the drive vertically mounted?

You might like to read my article on the subject.

Seagate's Seek Error Rate, Raw Read Error Rate, and Hardware ECC Recovered SMART attributes:
http://www.users.on.net/~fzabkar/HDD/Se ... R_HEC.html

[rry]

You might like to read my article on the subject.

Seagate's Seek Error Rate, Raw Read Error Rate, and Hardware ECC Recovered SMART attributes:
http://www.users.on.net/~fzabkar/HDD/Se ... R_HEC.html

An interesting, and informative read - Thank you fzabkar.

The drive is not vertically mounted - what is the significance of the orientation - airflow circulation?

[fzabkar]

The drive is not vertically mounted - what is the significance of the orientation - airflow circulation?

ISTR that a particular Seagate user reported seek errors (or high fly writes?) when his drives were vertically mounted but not when horizontally mounted. I can't remember which model it was, though. This was in Seagate's old forum, so the thread is no longer viewable.