I have a SAMSUNG SpinPoint T166 series SATA hard drive. I have been getting some worrying output from smartctl for this drive, including

199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       651


SMART Error Log Version: 1
ATA Error Count: 531 (device log contains only the most recent five errors)

However, SMART selftests (short and extended) all complete without error.

Periodically I get some IO lag and the following on dmesg:

ata3.00: exception Emask 0x10 SAct 0x0 SErr 0x280100 action 0x6
ata3.00: BMDMA stat 0x6
ata3.00: SError: { UnrecovData 10B8B BadCRC }
ata3.00: cmd 25/00:00:49:5a:a9/00:02:13:00:00/e0 tag 0 dma 262144 in
         res 51/84:6f:da:5a:a9/84:01:13:00:00/e0 Emask 0x30 (host bus error)
ata3.00: status: { DRDY ERR }
ata3.00: error: { ICRC ABRT }
ata3.00: hard resetting link
ata3.01: hard resetting link
ata3.00: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
ata3.01: SATA link down (SStatus 4 SControl 300)
ata3.00: configured for UDMA/133
ata3: EH complete

When this happens the UDMA_CRC_Error_Count goes up.

I'm fairly convinced this means my drive is broken. However, if you visit Samsung's website, you are greated with the following text:

Before you return your HDD, please run the SAMSUNG Diagnostic Program. If the defect is confirmed by the SAMSUNG diagnostic program, please follow the directions below

I gave the samsung utility a try (via the Ultimate Boot CD, an extremely useful tool) and after a long surface read/write scan it could find no errors.

I'm almost certain that this drive is not to be relied upon and that I should try and get it replaced by Samsung. I'm also somewhat pessimistic as to how easy this will be since their tool gives the drive the all-clear.

This would be drive death #5 for me in the last 12 months. Actually, drive death #5 ever for me.