All of a sudden getting XFS metadata I/O errors and on other occasions md disk read errors

[jbat66]

Been a reliable system up until a few days ago. Then started getting XFS metadata I/O errors and on other occasions md disk read errors per the console. I do a reboot, Get a warning message that several disks had read errors, but all are OK now, and parity check started. After 16 hours of check, it comes back good. and then stays running for a few days, until the next time. Generaly when this happens, no one is useing the shares, nor the dockers (of course they are still running in the background). While I have VM's they are not running. Upgraded to 6.9.2 a month or two ago, never had a problem on the previous version of unraid. 6.8.x

 

I have purchased a replacement drive, thinking it was disk2, but every time it 'seems' to be several disks with read errors.

Unraid version: 6.9.2
ASUSTeK COMPUTER INC. PRIME B450M-A, Version Rev X.0x
American Megatrends Inc., Version 1201
BIOS dated: Thu 25 Apr 2019 12:00:00 AM MDT
Mem: 16 Gig DDR4

CPU: AMD Ryzen 3 2200G with Radeon Vega Graphics @ 3850 MHz

thor-diagnostics-20220201-1851.zip

[JorgeB]

Feb  1 03:52:37 Thor kernel: ahci 0000:02:00.1: AHCI controller unavailable!

 

Problem with the onboard SATA controller, quite common with some Ryzen boards, look for a BIOS update and if that doesn't help best bet is to use an add-on controller (or a different board).

[jbat66]

5 hours ago, JorgeB said:

Feb  1 03:52:37 Thor kernel: ahci 0000:02:00.1: AHCI controller unavailable!

 

Problem with the onboard SATA controller, quite common with some Ryzen boards, looks for a BIOS update and if that doesn't help best bet is to use an add-on controller (or a different board).

Thank you, I will give that a try.