I recently pulled a bunch of drives (mostly WD RED) out of inventory to build up some RAID units. On testing (DLG) several of them reported bad sectors. Not tons, and normally nothing that I would not put into service. However, I cannot get the units to reallocate the sectors.
If I run a read surface test, it shows up with a number of bad sectors, and SMART shows “current pending errors count” for the sectors. If I then do a full surface erase, the pending goes to zero, and the “uncorrectable error count” remains at what the old “pending” count was - but the “Reallocation Sector Count” stays at zero. If I then run another surface read test, I am right back with bad sectors.
Not knowing the internal algorithm, it sounds like sectors that are “weak” - they will write okay, but won’t read correctly a couple of hours later, meaning that they never get reallocated.
Our DLG Diagnostic utility will normally scan the drive, and if it finds bad sectors, it will give an error and ask if you want them replaced. If you say to fix, then it will rescan and replace the bad sectors. If after it rescans, there are no other issues, then the drives should be fine to use. You won’t see any indications of sectors being replaced. However, if there continues to be bad sectors showing up, then you should probably replace the drives.
Unfortunately, in spite of the HDs apparently only having relatively few bad spots, only one of them “fixed.” When I told them to repair the drive, I almost immediately got a message about “Unable to repair.” [In spite of the fact that there appear to be zero sectors reallocated.]
Some of these are past their warranty date. The rest are still in warranty, so I guess I will return those that are beyond help.
