Beelink Me Mini drives disconnecting

TrueNAS General
9

Received my Beelink ME Mini less than a week ago.

I know this forum is mainly for TrueNAS users, but since this place has the biggest discussion thread about this device, I figured it’s the best place to post.

Hardware setup

  • Beelink ME Mini
  • 6 × Samsung 990 Pro 2TB (configured as RAIDZ)
  • Proxmox installed on internal eMMC (I know this isn’t recommended, but I don’t think it’s the root of the problem)

Issue

The system consistently loses the ZFS pool under semi-heavy load (e.g., ~300 MB/s copy from an external SSD to the pool).
I can reproduce this every time.

When the issue occurs, several NVMe drives drop out one by one, and the pool is suspended.

What I’ve tried so far, but none of the following helped:

  • Kernel parameters:
    nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off
  • Disabled Suspend and Hibernate options (can’t really remember the full name of parameter rn) for PCI devices in BIOS
  • Disabled Fast Boot (well, this can probably help to correctly boot with all 6 SSDs, but eventually the system will lose drives or even reboot)
  • Forced fans to 100% speed

Observations:

  • Highest recorded SSD temperature during copy: 65 °C
  • CPU average load per core: ~70%

Log when the pool dies:

nvme nvme2: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0xffff
nvme nvme2: Does your device have a faulty power saving mode enabled?
nvme nvme2: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug
nvme 0000:05:00.0: Unable to change power state from D3cold to D0, device inaccessible
nvme nvme2: Disabling device after reset failure: -19
.....
same for nvme5, then nvme1, and so on...
.....
WARNING: Pool 'masterpool' has encountered an uncorrectable I/O failure and has been suspended.

Based on [foxl] comment (can’t include links, sorry) the issue may not be the PSU itself, but the 3.3v line.

I can’t confirm this yet, but I did notice odd 3.3v readings in the BIOS. I can’t include images as well, but on 3.3v line it says +4.171V:

What options do we realistically have here?
If this is software-based, maybe a BIOS update or firmware patch could eventually fix it.
But if it’s hardware-based (and not something simple like an underpowered internal PSU), then I honestly don’t know what the path forward is.

But I really don’t want to give up after investing >900e in SSDs :confused: