Running GParted gives me an error that says
fsyncing/closing dev/sdb: input/output error
Using Gnome Disk Utility under the assessment section it says
Disk is OK, one bad sector
Clicking to format it to EXT4 I’m getting a message that says
Error formatting volume
Error wiping device: Failed to probe the device ‘dev/SDB’ (udisks-error-quark, 0)
Running sudo smartctl -a /dev/SDB I get a few messages that all say
… SCSI error badly formed scsi parameters
In terms of the physical side I’ve swapped out the SATA data and power cable with the same results.
Any suggestions?
Amazon has a decent return policy so I’m not incredibly concerned but if I can avoid that hassle it would nice.
If an HDD doesn’t work in every conceivable way out of the box it must be an RMA. These devices are too delicate to settle for anything less than flawless on arrival.
I would return it, but if you are curious you can try some of the following to get experiencing identifying bad disks.
You could try a different computer or controller to be sure.
If you can get some writes/reads to work, you can use badblocks or dm-crypt: https://wiki.archlinux.org/title/Badblocks#Alternatives
Badblocks will write known data to disk then read it to verify its good. If the disk is malicious, this can be faked. badblocks is also a little slow.
Using dm-crypt in the wiki will write zeros through dm-crypt which will result in random noise being written to disk, then compare with zeros to verify reads are good. This can not be faked easily since the zero stream is encrypted as it is written to disk.
Ive never had an HD arrive DOA, but one amazon new drive i ordered did that.
It was frustrating to diagnose cause that was the last thing I believed could be wrong.
if it’s functioning unexpectedly there should be signs it’s failing in the SMART data.
