Hi

What is the collective experience with trying to fix HDDs issues like:

 

Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error

# 1  Short offline      Completed: read failure      90%      3136        2916616800

 

My experience is the tools from the HDD manufactures can't fix this stuff - the drive it toast.  Am I correct?  If not, any tools recommended for this kind of thing - HDD manufacture's tools, spinrite, etc?

 

Thanks for the info

disk4_smartctl_report.txt

If you remove it from the unRAID server and install it in a windows pc then you can use the Western digital diagnostic tool to test it. What I have done in the past is to do a couple of passes where each pass consists of a full read test followed by a full write with zeros. The utility will tell you if it is having problems completing the test and if there's an issue you can start the rma process.

 

Stephen

Actually I created a DOS boot flash drive with all of the tools so I don't have to mess with windows.  I personally am getting sick of the lack of quality in these hard drives.  I have 3 that I am RMAing this week alone. They seem to last less than a year and my system is only used occasionally.  I guess the only thing positive is the RMA process is rather easy to do.

Hi

What is the collective experience with trying to fix HDDs issues like:

 

Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error

# 1  Short offline      Completed: read failure      90%      3136        2916616800

 

My experience is the tools from the HDD manufactures can't fix this stuff - the drive it toast.  Am I correct?  If not, any tools recommended for this kind of thing - HDD manufacture's tools, spinrite, etc?

 

Thanks for the info

I suspect you'll never be happy.

 

The read failure is reflected in ONE sector that is pending re-allocation.  You must have issued a request for a "short" smartctl test.  Both short and long test abort of the first un-readable sector they find.    That one sector is now marked for re-allocation the next time it is written to. 

 

Most modern disk have several thousand spare sectors.  That disk is nowhere near bad, and if the manufacturer were to check, would not be considered as a candidate for an RMA.

 

There are no parameters even close to their associated failure threshold.

 

Good luck.  It is not a bad disk. 

 

Oh yes, one more thing... there is nothing you can do to "fix" the un-readable sector.  All you can do is use a utility like the preclear_disk.sh script to fully test your disk.  If you see it identify un-readable sectors, and the number of un-readable sectors increases every time you run the utility (or when the number steadily grows to over several hundred), then it is time to RMA the drive.

 

Joe L.

Joe, you are probably correct on this one.  The extended test from WD is still running but the short test showed no error.  The drives I am RMAing this week are a dead 2TB WD black and the two 2tb seagates that issue the error code required to RMA them.  Maybe it's just par of the course. 

Joe, you are probably correct on this one.  The extended test from WD is still running but the short test showed no error.  The drives I am RMAing this week are a dead 2TB WD black and the two 2tb seagates that issue the error code required to RMA them.  Maybe it's just par of the course.

I will not say the disks are all reliable.  I've got one here that failed as brand new and I've got to RMA it, and another that has had several hundred re-allocated sectors that also needs to be sent back.  From what I've been seeing, about 1 out of 5 disks has issues.

 

A few bad sectors will not qualify a disk for an RMA for many manufacturers.  A completely dead disk certainly will qualify.

 

Joe L.

Joe, as your note said, that drive wasn't bad.  The WD DLGDIAG v5.19 tool said it repaired the drive after 6 hours of the extended test.  Now, with this drive and my others, after the RMAs come back, I will have to setup a 2nd server. Thanks