Hard Disk: SMART Scans (Seagate)

I do operater multiple hard disks as part of Network Attached Storage (NAS) and server disks. Especially at NAS systems, regular SMART tests are recommended. Smart stands for Self-Monitoring, Analysis and Reporting Technology.

SMART collects and analyzes many attributes of a drive’s operation. The most common SMART attributes for HDDs include:

1. Reallocated Sectors Count
   • Indicates bad sectors that have been remapped to spare areas.
   • High or increasing numbers suggest drive degradation.
2. Spin-Up Time
   • Measures how long it takes for the spindle to reach operational speed.
   • Slow spin-up time may indicate mechanical issues.
3. Power-On Hours
   • Tracks how many hours the drive has been powered on.
   • Useful for determining drive age and wear.
4. Start/Stop Count
   • The number of times the drive has been powered on and off or put into standby.
5. Temperature
   • Drive’s operating temperature; high or fluctuating values can shorten lifespan.
6. Seek Error Rate
   • Frequency of errors when the drive is positioning the read/write heads.
   • High rates may indicate mechanical wear or firmware issues.
7. Read Error Rate
   • Tracks hardware-level read errors.
8. Spin Retry Count
   • The number of times the drive had to retry spinning up.
9. Current Pending Sector Count
   • Sectors waiting to be reallocated due to read errors.
   • Can predict imminent failure.
10. Uncorrectable Sector Count
    • Number of sectors that cannot be recovered using error correction.

How to read SMART attributes (Seagate)?

Synology provides automatic SMART tests, which result in a table such as below:

Seagaute publishes supported attributes as part of Seagate SMART Attribute Specification (or see document below). Certain websites, such as disktuna.com, provide calculators, which convert RAW-Values into attributes, e.g.:

• Raw Error Rate (1)
  e.g. 243311455 (Raw Data) equals to 0 errors in 243.311.455 operations.
• See Error Rate (7)
  e.g. 201692307 (Raw Data) equals to 0 errors in 201.692.307 operations.
• ECC On the Fly Count (195)
  e.g. 243311455 (Raw Data) equals to 0 errors in 24.331.455 operations.

Attribute Name	No.	Normalized Attribute Range	Attribute Threshold	Status Byte Value	Pre-Fall Warranty	Notes
Raw Error Rate	1	1-166	6 Near Line=44	0Fh	Customer Specific	Updated online from Avalanche Onward
Spin Up Time	3	1-100	0	03h	No	No threshold
Start/Stop Count	4	1-100	20	32h	Yes
Retured Secotrs Count	5	1-100	36	33h	Yes	64% of guaranteed available spares used
Seek Error Rate	7	1-100	30	0Fh	Yes	Threshold of 30
Power On Hours	9	1-100	0	32h	No	No threshold
Spin Retry Count	10 Ah	1-100	97	13h	Yes
Drive Power Cycle Count	12 Ch	1-100	20	32h	Yes	No threshold
Attribute ID 184: Reported IOEDC errors	184 B8h	1-100	99	32h	No
Reported Uncorrectables	187 BBh	1-100	0	32h	No
Command Timeout Count	188 BCh	1-100	0	32h	No
High Fly Writes	189 BDh	1-100	0	3Ah	No
Air Flow Temperature	190BEh	-127 to 127	45	22h	No
Temperature	194 C2h	-127 to 127	0	22h	No	Degrees C. New in Vail.
ECC On the Fly Count	195 C3h	1-166	0	1Ah	No
Pending-Sparing Count	197 C5h	1-100	0	12h	No
Uncorrectable Sectors Count	198 C6h	1-100	0	10h	No
Ultra DMA CRC Error Count	199 C7h	1-200	0	3Eh	No	R-Errs in SATA drives
			defaults	defaults	Bit 0 of Status byte indicates if the attribute is pre-fail Warranty. 1 is yes.

Please find a detailed documentation as of Seagates published document from 2011:

Types of SMART Tests

There are usually three kinds:

1. Short Test
   • Quick check of major components (~1-5 minutes).
   • Good for routine monitoring.
2. Extended (Long) Test
   • Full surface scan and in-depth test (~tens of minutes to hours).
   • More reliable at detecting failing drives.
3. Conveyance Test
   • Designed to detect damage incurred during transport.
   • Shorter than extended, but rarely used by consumers.