Hello! Long time Unraid users and I have never encountered this issue before. A recent scheduled parity check reported 8 sync errors. According to the Unraid scheduler settings, this was a non-correcting parity check. I would really appreciate some guidance to understand the steps I need to take. Thanks! Diagnostics attached. 

tower-diagnostics-20231031-2142.zip

Edited November 1, 20232 yr by JPunraid

Log is full of spam so cannot see the beginning, run another non correcting check to see if it finds the same errors, if yes correct them.

Thanks very much for the assistance.

 

I ran another non-correcting parity check as instructed. New diagnostics attached. 

 

I looked through the logs and it looks like the parity check found the errors in the same sectors. The only error I can't be certain of is the very first error, as it was logged ahead of all the log spam. I noted the sectors from each check below. 

 

I also just ran short smart tests on all disks and no errors were thrown. 

 

 

Is there a way to determine which files these errors are associated with? Or is there a way to determine whether the errors are on the array drives versus the parity drive?

 

Thanks so much for the assistance!

tower-diagnostics-20231104-0903.zip

12 minutes ago, JPunraid said:

Is there a way to determine which files these errors are associated with? Or is there a way to determine whether the errors are on the array drives versus the parity drive?

No practical way to find what files might be affected (if any).

 

Since the write order is data drives before parity then as long as you do not have a hardware error you have to assume it is the parity drives that are out of sync.   The remedial action is to run a correcting parity check.   This should report the same number of ‘errors’ but it corrects them so that subsequent checks should find 0 errors.

Understood, thanks very much for the input and insight. 

 

In the last year or so, I have been generating md5 hash files for anything I add to my server. Coincidentally, the server was off for most of the month of October, so I don't actually have that many new files to check. I run parity checks every month, so, presumably I should be able to check the new files that have been added to the array since the last known good check.

 

Either way, I will take the suggested action and run a correcting parity check and then a non-correcting check to validate that things were fixed. 

 

Appreciate all of the help on this.  

 

 

Edited November 4, 20232 yr by JPunraid

If you have checksums that is the way to confirm whether any files that are corrupt.

 

If you are using btrfs or ZFS as the format of drives in the array these format have built-in check-summing of files.   However XFS is more frequently used on array drives as it is more performant and less prone to file system corruption.

Yeah, I have been using XFS. 

 

I'll spend the time validating the checksums on the files, it shouldn't take too long. If I find no errors, I'll feel better about running the correcting parity check. 

I verified the checksums of every file I could find that was dated with a creation date after my last known good parity check. All checksums matched. 

 

I also ran a correcting parity check, and on this run the same 8 errors with matching sector locations were found and corrected. 

 

I just kicked off a non-correcting parity check. 

Non-correcting parity check just completed, finding zero errors. Thanks very much all. Marking this as solved. I appreciate the help.