Need advice about possibly defective HDD

[68k2x86]

After buying a 4TB WD HD, I captured the SMART stats, ran badblocks (4 passes, read + write), and noted the SMART stats again, looking out for reallocated sectors.

I didn't get any reallocated sectors, but I did get 6 raw read errors and badblocks reported 4 bad blocks during the first pass.

I have seen mentioned that read errors can be caused by such things as electrical interference with SATA cables, so are these errors likely transient, or should I try to get a replacement (I think it has a 3 year warranty)?

 

[drescherjm]

When I get any problems in badblocks I then use the -p2 option to make sure it completes 2 entire runs of the 4 tests before I call it okay. If I see any more failure I always RMA the drive (well except for the firmware bug that I found on 2TB Samsung f4 drives a few years back).

 

[zfsgeek]

Try to run an extended smart test, if that fails too I'll definitely request a warranty replacement.

> smartctl -t long /dev/<>

 

[68k2x86]

Try to run an extended smart test, if that fails too I'll definitely request a warranty replacement.

> smartctl -t long /dev/<>

I actually did but since it came up clean I didn't include it in the original post. Does this change your recommendation?

 

[68k2x86]

When I get any problems in badblocks I then use the -p2 option to make sure it completes 2 entire runs of the 4 tests before I call it okay. If I see any more failure I always RMA the drive (well except for the firmware bug that I found on 2TB Samsung f4 drives a few years back).

In other words, I should run bad blocks again, with an additional 8 passes, and if all is well, keep the drive?

 

[drescherjm]

In other words, I should run bad blocks again, with an additional 8 passes, and if all is well, keep the drive?

That is what I do at work.