Parity Device diabled and Disk offline

[Patmanduu]

The other day I noticed that one of my two, fairly new parity disks was disabled. I checked the SMART report and everything looked good so I proceeded with the following steps for the disabled parity disk:

 

[*]Stop the array

[*]Unassign the disk

[*]Start the array.

[*]Stop the array

[*]Assign the disk

[*]Start the array and parity build should start

 

After about 20 hours the rebuild finished successfully. Then a few hours later the same parity disk dropped disabled again, AND one of the older array disks dropped into a disabled/emulated state. The affected array disk was pretty old and I had a new one on hand so I just replaced it (as per https://wiki.unraid.net/Replacing_a_Data_Drive). I also checked my SATA cables and reseated the LSI card.

 

Again it took around 20 hours to rebuild and all disks came back online briefly. Then a little later the same problem occurred, parity disk 2 disabled and the brand new array disk in emulated state. Both these disks are new Ironwolf NAS drives with clean SMART reports (at least in my limited interpretation).

 

I repeated the above after swapping around some SATA cables to check if they were the cause but no luck there. I can't swear that the affected array disk was emulated right from the start, but since it has been replaced I am not sure what is going on. Any help would be greatly appreciated! Diagnostics attached.

 

syslog is showing both read and write errors to "disk29" though I am unsure which disk that is.

 

Parity 2's disk log shows several:
May 19 15:21:52 Spring kernel: blk_update_request: I/O error, dev sdj, sector 722319680 op 0x0:(READ) flags 0x0

 

spring-diagnostics-20220519-1556.zip

 

Edited May 20, 2022 by Patmanduu

[JorgeB]

See here:

 

 

[Patmanduu]

Thank you @JorgeB. The two affected drives are indeed both 8TB Ironwolf ST8000VN004, and are both connected to my LSI card, so it is likely caused by the issue you linked.

 

Before I go modifiying the disk settings however, I decided to try moving the affected disks off the LSI and on to the motherboard SATA, which should confirm if that is the problem. I have a total of 12 disks (only three of which are ST8000VN004) and 8 SATA ports on the MB, so they do not need to be connected to the LSI.

 

So far, the array disk is back online and the parity 2 rebuild is progressing normally.