I recently bought a Lincstation N2 and 4x4TB Kingston KC3000 drives to go with it.
I’m having real bad issues with it though. When doing heavy reads, after a few seconds randomly one of the drives display as removed and pool health is degraded. Sometimes random drives also show I/O errors, it’s all random which drive gets removed and I/O errors.
I get errors like this in dmesg
[ 2834.725144] nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10
[ 2834.725149] nvme nvme1: Does your device have a faulty power saving mode enabled?
[ 2834.725152] nvme nvme1: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug
[ 2834.761172] nvme 0000:01:00.0: enabling device (0000 -> 0002)
[ 2834.761286] nvme nvme1: Disabling device after reset failure: -19
I’ve added quiet acpi_enforce_resources=lax nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off to grub and last time after testing I got the I/O errors.
Any suggestions? Is the N2 device faulty? It seems very unlikely that all 4 drives would be bad since they are brand new..
We have seen similar reports from the Beelink Mini. There it was an inadequate power supply. To address 2x sata and 4x nvme, there also is some pci switching going on to achieve that with 6 pcie lanes of the n100.
I would systematically run a pool with 1, 2, 3 or 4 drives and see how far you get.
Also create a pool with the included unraid license to see if it is ZFS simply putting to much strain on the system.
Can you try testing with a default hardware and software setup. If it fails then I would contact Lincstation for support. Is there an approved list of NVMe? Maybe their list is lower powered devices?
I’ve tried with unraid aswell as it’s included get errors there aswell. Can’t see a list but I found once site they’ve published themself recommending this drive