[SOLVED] Parity errors at the third rebuild

[Marius1085]

Hey,

 

I searched the forum for fitting answers but cant seem to find any for my problem.

I already changed the SATA cables, and precleared 2 times and ran an extended smart but UNRAID is still throwing errors at my face.

 

So I dont know whats happening. I rebuild my parity drive for the third time now but still get parrity errors.

First and second time I got parity errors when I wrote from my cache drive. On the third time the first write from cache worked fine but before my second scheduled cache write (I set mine to 0:00 everyday) I got parity errors again. I want to have a parity drive for extra security but Im at the point where I just want to change that to a data drive. 

I dont know where but I saw someone using turbo write? so in disk setting he changed turnable from auto to reconstruct write. Would this help?

 

Please tell me what Ive done wrong and how I can fix it. 

 

I added the diagnostics logs.

 

Thanks for your help

quantumbreak-diagnostics-20200625-0017.zip

[JorgeB]

It's a problem with the onboard SATA controller:

Jun 24 23:08:02 QuantumBreak kernel: ahci 0000:02:00.1: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x44c2b16d89f81e00 flags=0x0010]
Jun 24 23:08:02 QuantumBreak kernel: ahci 0000:02:00.1: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x44c2b16d89f81e80 flags=0x0010]

This is quite common with Ryzen servers and there are reports that upgrading to v6.9-beta helps because of the newer kernel.

[Marius1085]

Ok, Ill try upgrading and see how that works out. 

 

Ill let you know if that solved my problem.

[Marius1085]

Another question. As the parity had some failures I also saw some errors on the main page for some of the other array disks.

For example parity had 1300 errors and 3 other disks shared these 1300 errors. 

This only happened when cache wrote to array while parity was active.

 

What do these errors say? Could it be that some files are corrupted? If so can I see which files are corrupted?

[JorgeB]

13 minutes ago, Marius1085 said:

Could it be that some files are corrupted?

All files already on the disks should be fine, if there were any files being written when the controller crashed those are likely corrupt.

[Marius1085]

Can I see which files were corrupted?

Those were all media files, so nothing too bad but if I could see which are corrupt that would be cool.

 

As of now everything is fine and parity didnt show a single error. Ill tell you again in a few days to see if my parity is working fine.

[JorgeB]

8 hours ago, Marius1085 said:

Can I see which files were corrupted?

Only if you have pre-created checksums for all files (or were using btrfs).

[Marius1085]

Thank you very very much johnnie. 

 

The errors didnt show up and everything works fine.

 

Seems to be the update to a newer kernel is all it takes.