I've been reading the forums trying to find the answer to my question but wanted to ask to make sure I am acting correctly.  I have a ups on my system but two nights ago after a power outage, the ups did not cleanly shut down my Unraid box.  I adjusted the setting on the ups to power down with 20% remaining instead of the 10% that it was configured for.  Upon power up it started a non correcting parity check as expected.  Disk 4 in my system started getting errors in the main GUI about an hour into the parity check.  The party check finished with 0 sync errors, but my drive shows 103,225 errors in the GUI.  None of the other drives had any issue.  Could this have to do with the power outage or was it probably just a coincidence that it happened on the parity check afterwards?

 

My big question are how did i need to proceed and is my data on that drive still in good shape.  I'm just looking for reassurance.

 

I have attached the SMART report for the drive and syslog here.  The Reallocated_Sector_Ct and Reallocated_Event_Count values both went up while the parity check was running.  

 

Having had 0 sync errors in the parity check can I safely replace the drive with a new one or is that even necessary at this time?

 

edit:  Log files removed

 

 

Edited December 16, 20205 yr by toopty

Data should be fine since Unraid successfully wrote back the failing sectors (assuming parity was valid), disk doesn't look very healthy though.

The parity check was the auto started un-correcting one.  How do you know that it wrote the data back successfully?  I'd like to be able to verify that in the future on my own.  The process had 0 sync errors so I think I'm confusing myself on parity errors vs read errors, or just don't understand the way Unraid is reporting to me.

 

I agree about the drive looking like it's in bad shape, but wanted to get a second set of eyes on it to make sure I didn't replace anything unnecessarily.   It's one of my newer 3 TB drives but won't be any issue to replace.  As long as I can do it safely.

 

As far as replacing the drive, I just need to remove it from the array and let it rebuild to a replacement drive.  Will the existing read errors affect that process?  I have a known good 3tb that I removed  in the past to upgrade to a larger drive that will replace it just fine.

1 hour ago, toopty said:

How do you know that it wrote the data back successfully? 

Because the disk wasn't disabled, any write error Unraid would disable the disk.

 

1 hour ago, toopty said:

As far as replacing the drive, I just need to remove it from the array and let it rebuild to a replacement drive.  Will the existing read errors affect that process?

Just do a standard replacement, as long as no more errors on a different disk it should be rebuilt correctly.

Thank you so much for the advice and reassurance.  You guys are always helpful and quick to respond.