The day following an update to 6.12.6 I received a warning about a power event that occurred (strange because the box is connected to a UPS) and then was followed by one of my disks being marked as disabled due to a failed write. After reboot the drive did not get assigned as normal.  I checked the cables for both drives and I ran a SMART check on the disk and it completed without any errors. Both drives show as healthy.

To test if the parity was still ok I added a new config, started in maintenance mode and ran a non correcting parity check. 
It passed parity, but it came back with a large number of errors. 
Both disks show as healthy and I can mount the data for the emulated drive. 

What is the best option to move forward? Should I just replace and rebuild the failed drive from parity? 


Currently only have two drives : 

Parity
disk1 - (Failed)

Latest Diagnostics attached. 

tower-diagnostics-20231217-1128.zip

54 minutes ago, 0xj4r said:

To test if the parity was still ok I added a new config, started in maintenance mode and ran a non correcting parity check. 

Not sure I follow, a new config should not be done if there's a disabled disk, please confirm if that is really what you did.

 

55 minutes ago, 0xj4r said:

It passed parity, but it came back with a large number of errors. 

Sync errors or disk errors? Diags posted show sync errors.

 

19 hours ago, JorgeB said:

Not sure I follow, a new config should not be done if there's a disabled disk, please confirm if that is really what you did.

 

Yes, after the reboot the drive was not recognized. This has happened in the past when an improper shutdown has occurred. Assigning the drives in the correct location and using new config has always worked for these situations. From what I could, tell it looked like the write error happened due to whatever power abnormality happened with the disk. Since the SMART tests seemed to return that the drive was fine I thought perhaps it could be repaired instead of having to rebuild the entire drive. 
The array was started in maintenance mode and after checking valid parity checkbox. 
 

 

19 hours ago, JorgeB said:

Sync errors or disk errors? Diags posted show sync errors.

 Sync errors I believe. Parity was labeled as valid and the parity check passed. 

Is there anything else that would be helpful to provide? 

You usually should not do a new config after a disk gets disabled, but after an unclean shutdown a few sync errors are normal, just run a correcting check.

Thank you for all of your attention and advice. 

I just wanted to clarify a few more details before running a correcting check since the data on this disk is important:

 

- There were over 400,000 sync errors found during the non-correcting parity check that was marked valid/successful. 
- There were many unclean shutdowns since my last parity check due to power loss issues before receiving the new UPS . I'm hoping that this is just due to those.  

Since the non-parity drive (disk1) was the disk disabled will correcting parity to match disk1 potentially corrupt what is currently in the parity drive (parity 1) or is that not how the data is propagated for data writes? 

Also, what should I do in the future if a non-parity disk gets marked as disabled? I have looked in the docs and on the forum, but still unclear on this. 
 

If you have been doing new config after a disk gets disabled it's normal that parity will become out of sync, and there will be more errors each time you do it.

 

16 minutes ago, 0xj4r said:

Also, what should I do in the future if a non-parity disk gets marked as disabled? I have looked in the docs and on the forum, but still unclear on this. 

Depends on why the disk got disabled.