Dear,
I am currently facing a (from my perspective) strange issue. After switching hardware, I am currently refilling my HDDs with the data that it should have.
Unfortunately, I am getting into issues here.
When I start to copy the data (either via SMB or via SFTP), everything is running smoothly. The network speed is about 100 - 115 MB/s and the files are being
1. written and
2. evenly distributed on my mergerfs-pool (consisting of two HDDs (sdb and sdc)
What I can see in the dashboard is that there are a lot of Soft-IRQs, like this
After "some" data being transferred (100 GB to 2,5 TB), at least one of the HDDs is falling into RO mode, which obviously stops any transfer.
As soon as this happens, I do a reboot via the GUI (which is still available). After the reboot the data transfer can start again.
My current setup:
NUC11ATKPE (Pentium Silver N6005)
16 GB RAM
1 M2-SSD (/dev/nvme0n1p2)
2 HDDs (/dev/sdb, /dev sdc)
1 SSD (/dev/sda)
connected via a 4 bay HDD enclosure and connected via USB.
My idea is that there are some drivers for the USB ports are missing, since the HDDs were working without any issue on a RaspPi v4.
I have enclosed roughly 15 minutes of syslog events - hopefully this helps, otherwise, any idea is highly appreciated :-D.
Thank you for pushing me in the right direction.
Best regards
Gilhin