Hi,

 

I have my unRAID set up to regularly email me the health status of my array and today I noticed that one of my drives is coming up with "(disk has read errors) [NOK]"

I am not sure what to do from here.

I am trying an extended SMART test to see if I can find out more specifically what is going wrong but is it as I fear and it needs physically replacing? Or can this be configuration/software related?

 

Please advise on any logs I need to post in order to provide detailed information as I am not sure where to look myself.

 

Thanks in advance for any help !

While you're waiting for the SMART test to complete read this: https://lime-technology.com/forum/index.php?topic=39257.0

Thank you for the prompt response.

 

Please see attached the diagnostics that I downloaded and SMART details for the affected drive

whitecorsair-diagnostics-20160808-1910.zip

WDC_WD30EZRX-00MMMB0_WD-WMAWZ0077440-20160808-1854.txt

Let the extended SMART finish but it looks like a disk with bad sector(s).

Hi, the SMART check completed "without errors" It was the extended test which I ran.

I have since stopped and restarted my array to reset the errors counter in unRAID and no more errors have popped up since, meaning my regular status emails are coming through as "all ok"

 

Should I be concerned that there is an area of my disk which is bad and it will only error out when trying to access it? Or is the bad sectors within tolerance of the SMART test somehow? Is there any further testing I should do on the drive to ensure it isn't on its way to failing on me?

 

Thanks again for your help

WDC_WD30EZRX-00MMMB0_WD-WMAWZ0077440-20160809-1310.txt

Disk looks to be OK for now, the fact that after the read errors it wasn't disabled by unRAID means that the data was successfully re-written to the affect sectors, still read errors are never a good sign, but it's difficult to predict if it's going to fail soon, I would definitely replace it if you get more read errors in the future.