Hello, my order : #22337 I try to use ME mini as a NAS. Installed Truenas Scale 25.04.2.1. The ME mini is connected to 10gbe switch. 1/ NVME_IOCTL_ADMIN_CMD: Input/output error. First i tryed to use it with 3 WD_BLACK 2TB SN850 NVMe (model : WDS200T1X0E-00AFY0). Then i bought 3 Crucial T500 SSD (model : CT4000T500SSD3). Now i have a 2 WD_BLACK (slots 2, 3) and 3 Crucial (slots 4,5,6) installed. My pool is composed of 3 Crucial + 1 WD_BLACK. Have this error with all my WD_BLACK M.2s : Controller failed: NVME_IOCTL_ADMIN_CMD: Input/output error. Device: /dev/nvme0n1, failed to read NVMe SMART/Health Information. SSD disconnected, pool degraded, after reboot i could repair pool replacing failed drive with other drive (resilver, scrub). But then other drive diconnects the same way in an other slot some time later. Also i see this in sudo cat /var/log/messages : Aug 13 01:27:06 mini kernel: nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0xffff Aug 13 01:27:06 mini kernel: nvme nvme1: Does your device have a faulty power saving mode enabled? Aug 13 01:27:06 mini kernel: nvme nvme1: Try “nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off” and report a bug Aug 13 01:27:06 mini kernel: nvme nvme1: Disabling device after reset failure: -19 2/ Tryed 5gbe USB wavelink ethernet adapter It disconnected after few minutes/hours of testing. 3/ Tryed 10gbe Ethernet card via M.2 adater in slot 1 It disconnected after few tests with iperf3. To test disks i disconnected ethernet 5/10gbe adapters, so the problem appears without anything connected (anything other than NVME themselves). CPU and Disks go to 60-80 degrees celcius quickly when there is activity. I opened the ME Mini to see if fan is working, it works but does not seem to do much. I do not hear it at all. Looks like my WD_BLACK drives are removed all the time few minutes after stop / wait 5 min / restart. Perhaps they use too much power and are not properly dissipate heat ? Perhaps the system is overheating quickly when there is activity ? Is there bios settings or something to boost the fan ? Help please :) Also i am in France, i bought from https://www.bee-link.com/ . So i hope there is a solution other then return back the device to US. And it would be great to have special forum entry for ME mini, because ME mini messages are difficult to find.

found this command and executed, seems better, will put it to work and see if become stable : midclt call system.advanced.update '{"kernel_extra_options": "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off"}' Related threads in Truenas forum : https://forums.truenas.com/t/beelink-me-mini-drives-disconnecting/50143 https://forums.truenas.com/t/truenas-beelink-me-mini-10gbps-via-aquantia-corp-aqc107-nbase-t-ieee-802-3bz-ethernet-controller-aqtion-rev-02/50156

One more thing (M.2 drives disconnection problem with Truenas) : use command above + desactivate suspend options in bios.

Me mini + Truenas scale : 5 gbe / 10 gbe not working or with degraded performance. There is driver problems because Truenas Scale has not the most recent linux kernel, poorly supporting some devices, and there is no way to install more recent linux drivers. There is also cooling problems witn 10gbe card used with M.2 adapter/extender, you need add better cooling. Tryed one cheap 10gbe adapter, sent it back to amazon, will try an other one.

Finally made 10gbe work : Proxmox > 10gbe tuning in Proxmox > Truenas VM https://forums.truenas.com/t/truenas-beelink-me-mini-10gbps-via-aquantia-corp-aqc107-nbase-t-ieee-802-3bz-ethernet-controller-aqtion-rev-02/50156/10?u=dogelab

doge This sounds very interesting to me and I’m considering also setting up Proxmox on this Me Mini with Truenas as a VM. Is it working well for you and any issues or downsides to be aware of compared to using just Truenas straight? Did you install Proxmox on the 64gb eMMC or on one of the m2 nvmes?

Hello, with Truenas you have locked kernel and linux system, there is very few room to try to test/tweak things when you have a problem. Thats’s why i try run truenas on top of proxmox (proxmox is basic debian, with full shell and linux commands allowed). Proxmox also has more recent kernel with more recent drivers. With proxmox you can complile / install / run a driver, with Truenas you can’t. With Proxmox 9 install when installed it it did not permit me to fromat my USB M.2 drive with default EXT4, so i user ZFS, the only that worked. Unfortunatly after last kernel/proxmox update, when i boot it can’t auto select/mount zfs, so i need to type “zfs import rpool” every time i restart the system. Proxmox 9 is too recent and has many problems. Never had update that broke the boot with Proxmox 8. On the EMMC i have my Truenas direct install (1st test install), on the USB M.2 i have Proxmox (2nd test install) : i can boot on both. In the Beelink Me mini i have the 10gbe M2 ethernet card + 5 Crucial T500 4tb drives. In BIOS i desactivated power management, all my settings : https://drive.google.com/drive/folders/1qKy8X5LHsY8KTqGCWHt1RucNrNHCf933?usp=drive_link Unfortunatly now i am having problems with my 5 Crucial T500 M.2 4tb drives witch are going to D3cold state and break my RaidZ1 in tuenas. 2 of 5 M.2 drop from the system (they go to D3cold state and never come back to D0 state) : the only way to have them come back is power off completely the Beelink Me Mini, then very quickly they drop again. JayTwoCents has the exact same problem on windows : He says all M.2 drives witch have Physon E25 controller (not only the T500) have the same problem (in his tests with windows). He says this is related to the last Windows update. But i have the same problem (M.2 power off to D3cold and can not go back to D0 until full system power off) in Linux (debian proxmox 9). This leads me to think that Crucial T500 (CT4000T500SSD3) has bad firmware (version : P8CR013) that does not support all fonctions needed to bring back it from D3cold to D0. root@pxm:~# bash /var/lib/vz/snippets/nvme-firmware.sh [ ] Scanning PCI NVMe controllers… [*] Device 0000:03:00.0 driver=none desc=“03:00.0 Non-Volatile memory controller [0108]: Micron/Crucial Technology T500 NVMe PCIe SSD [c0a9:5415] (rev 01)” vendor=c0a9 device=5415 [ ] Device 0000:03:00.0 appears to be in D3cold. [*] Attempting to wake 0000:03:00.0 from D3cold… [!] 0000:03:00.0 remains in D3cold — skipping [ ] Device 0000:04:00.0 driver=none desc=“04:00.0 Non-Volatile memory controller [0108]: Micron/Crucial Technology T500 NVMe PCIe SSD [c0a9:5415] (rev 01)” vendor=c0a9 device=5415 [*] Device 0000:04:00.0 appears to be in D3cold. [ ] Attempting to wake 0000:04:00.0 from D3cold… [!] 0000:04:00.0 remains in D3cold — skipping [*] Device 0000:05:00.0 driver=none desc=“05:00.0 Non-Volatile memory controller [0108]: Micron/Crucial Technology T500 NVMe PCIe SSD [c0a9:5415] (rev 01)” vendor=c0a9 device=5415 [ ] Adding PCI ID c0a9:5415 to nvme driver new_id [*] Binding 0000:05:00.0 to nvme ==> 0000:05:00.0 (nvme0) MODEL “CT4000T500SSD3” FW “P8CR013” [ ] Device 0000:07:00.0 driver=none desc=“07:00.0 Non-Volatile memory controller [0108]: Micron/Crucial Technology T500 NVMe PCIe SSD [c0a9:5415] (rev 01)” vendor=c0a9 device=5415 [*] Adding PCI ID c0a9:5415 to nvme driver new_id [ ] Binding 0000:07:00.0 to nvme ==> 0000:07:00.0 (nvme1) MODEL “CT4000T500SSD3” FW “P8CR013” [*] Device 0000:08:00.0 driver=none desc=“08:00.0 Non-Volatile memory controller [0108]: Micron/Crucial Technology T500 NVMe PCIe SSD [c0a9:5415] (rev 01)” vendor=c0a9 device=5415 [ ] Adding PCI ID c0a9:5415 to nvme driver new_id [*] Binding 0000:08:00.0 to nvme ==> 0000:08:00.0 (nvme2) MODEL “CT4000T500SSD3” FW “P8CR013” [ ] Restoring 0000:03:00.0 back to none [*] Restoring 0000:04:00.0 back to none [ ] Restoring 0000:05:00.0 back to none [*] Restoring 0000:07:00.0 back to none [ ] Restoring 0000:08:00.0 back to none [*] Done. nvme-firmware.sh : script generated by Google Gemini to get firmware version for my drives #!/usr/bin/env bash set -euo pipefail # nvme-firmware.sh # Finds Crucial T500 NVMe devices, binds to nvme, reads firmware, restores original driver. # Handles D3cold and gracefully skips ghost/unpowered devices. log() { echo "[*] $*"; } err() { echo "[!] $*" >&2; } require_root() { [[ $EUID -eq 0 ]] || { err "Run as root"; exit 1; }; } require_tools() { for b in nvme lspci; do command -v "$b" >/dev/null 2>&1 || { err "Missing required tool: $b"; exit 1; } done } load_modules() { modprobe nvme >/dev/null 2>&1 || true; modprobe vfio-pci >/dev/null 2>&1 || true; } list_nvme_bdfs() { for d in /sys/bus/pci/devices/*; do [[ -f "$d/class" ]] || continue if grep -q '^0x0108' "$d/class"; then basename "$d" fi done } current_driver() { local bdf="$1" [[ -L "/sys/bus/pci/devices/$bdf/driver" ]] && basename "$(readlink -f "/sys/bus/pci/devices/$bdf/driver")" || echo "none" } unbind_from_driver() { local bdf="$1" drv="$2" local unbind="/sys/bus/pci/drivers/$drv/unbind" [[ -w "$unbind" ]] && echo "$bdf" > "$unbind" 2>/dev/null || true } bind_to_driver() { local bdf="$1" drv="$2" local bind="/sys/bus/pci/drivers/$drv/bind" if [[ -w "$bind" ]]; then if ! echo "$bdf" > "$bind" 2>/dev/null; then return 1 fi return 0 else return 1 fi } ensure_nvme_knows_id() { local vendor="$1" device="$2" if [[ -w /sys/bus/pci/drivers/nvme/new_id ]]; then log "Adding PCI ID $vendor:$device to nvme driver new_id" echo "$vendor $device" > /sys/bus/pci/drivers/nvme/new_id 2>/dev/null || true fi } bdf_to_nvme_ctrl() { local bdf="$1" for ctrl in /sys/class/nvme/nvme*; do [[ -e "$ctrl" ]] || continue if [[ "$(readlink -f "$ctrl/device")" == *"/$bdf" ]]; then basename "$ctrl" return 0 fi done return 1 } wait_for_ctrl() { local bdf="$1"; local timeout="${2:-20}" local i=0 while (( i < timeout )); do if bdf_to_nvme_ctrl "$bdf" >/dev/null; then return 0; fi sleep 0.2 (( i++ )) done return 1 } read_model_fw() { local ctrl="$1" local mn fr mn="$(nvme id-ctrl "/dev/$ctrl" 2>/dev/null | awk -F: '/^[[:space:]]*mn[[:space:]]*:/ {print $2}' | xargs || true)" fr="$(nvme id-ctrl "/dev/$ctrl" 2>/dev/null | awk -F: '/^[[:space:]]*fr[[:space:]]*:/ {print $2}' | xargs || true)" [[ -n "$mn" || -n "$fr" ]] || return 1 echo "${mn}|${fr}" } is_d3cold() { local bdf="$1" if [[ -f "/sys/bus/pci/devices/$bdf/power_state" ]] && grep -q 'D3cold' "/sys/bus/pci/devices/$bdf/power_state"; then return 0 fi if [[ -f "/sys/bus/pci/devices/$bdf/uevent" ]] && grep -q 'D3cold' "/sys/bus/pci/devices/$bdf/uevent"; then return 0 fi return 1 } try_wake_from_d3cold() { local bdf="$1" log "Attempting to wake $bdf from D3cold..." [[ -w "/sys/bus/pci/devices/$bdf/power/control" ]] && echo on > "/sys/bus/pci/devices/$bdf/power/control" 2>/dev/null || true [[ -w "/sys/bus/pci/devices/$bdf/power/wakeup" ]] && echo 1 > "/sys/bus/pci/devices/$bdf/power/wakeup" 2>/dev/null || true sleep 1 ! is_d3cold "$bdf" } main() { require_root require_tools load_modules declare -A ORIGINAL_DRIVER=() log "Scanning PCI NVMe controllers…" mapfile -t BDFS < <(list_nvme_bdfs) [[ ${#BDFS[@]} -gt 0 ]] || { err "No NVMe-class PCI devices found."; exit 1; } for bdf in "${BDFS[@]}"; do drv="$(current_driver "$bdf")" ORIGINAL_DRIVER["$bdf"]="$drv" vendor="$(printf "%04x" $(( $(cat /sys/bus/pci/devices/$bdf/vendor) )))" device="$(printf "%04x" $(( $(cat /sys/bus/pci/devices/$bdf/device) )))" desc="$(lspci -nn -s "$bdf" 2>/dev/null || true)" log "Device $bdf driver=$drv desc=\"$desc\" vendor=$vendor device=$device" if ! echo "$desc" | grep -qi 'T500'; then log "Skipping non-T500 $bdf" continue fi if is_d3cold "$bdf"; then log "Device $bdf appears to be in D3cold." if try_wake_from_d3cold "$bdf"; then log "Woke $bdf from D3cold." else err "$bdf remains in D3cold — skipping" continue fi fi ensure_nvme_knows_id "$vendor" "$device" if [[ "$drv" != "nvme" ]]; then [[ "$drv" != "none" ]] && unbind_from_driver "$bdf" "$drv" log "Binding $bdf to nvme" if ! bind_to_driver "$bdf" nvme; then err "$bdf did not respond to bind (likely absent/unpowered) — skipping" continue fi if ! wait_for_ctrl "$bdf" 20; then err "No nvme controller appeared for $bdf — skipping" continue fi fi ctrl="$(bdf_to_nvme_ctrl "$bdf" || true)" if [[ -z "$ctrl" ]]; then err "No /dev/nvme* controller found for $bdf — skipping" continue fi info="$(read_model_fw "$ctrl" || true)" if [[ -n "$info" ]]; then echo "==> $bdf ($ctrl) MODEL \"${info%%|*}\" FW \"${info##*|}\"" else err "Failed to read model/fw for $bdf ($ctrl)" fi done for bdf in "${BDFS[@]}"; do orig="${ORIGINAL_DRIVER[$bdf]}" [[ "$orig" == "nvme" ]] && continue log "Restoring $bdf back to $orig" unbind_from_driver "$bdf" nvme || true [[ "$orig" != "none" ]] && bind_to_driver "$bdf" "$orig" || true done log "Done." } main "$@" if(window.hljsLoader && !document.currentScript.parentNode.hasAttribute('data-s9e-livepreview-onupdate')) { window.hljsLoader.highlightBlocks(document.currentScript.parentNode); }

After having many problems with Truenas and Proxmox and my 10gbe+5 x Crucial T500, i decided to try to back to windows and try some Storage Spaces to achieve my 5 m.2 raid setup. My plan was to copy the PXE OS to une USB M.2, then use second USB M.2 to install windows (because EMMC is too small for updates). The PXE OS (setup to install to USB) did not detect the second USB and overwrite himself. So i tryed to install to EMMC. After many trial it finally installed (i had to cut my AdGuard, and not pluget internet so it let me create loca account). Once in Windows i discovered that my second USB M.2 is dropping or si not visible. So i tested id in my PC, and it works fine. Also i can see only 2 of my 5 internal Crucial T500 M.2. I finally understood that having 10gbe card + 5 internal M.2 + 2 USB M.2 simply draws too mych power. So i used USB hub with external power, and now my 2 USB M.2 are visible and work. Then i cloned the Windows installed on the EMMC (using DiskGenius) to the USB M.2, and rebooted, and it works fine. Now i have external USB Windows install with 2tb space and i can do Windows update. There also wad a peripheral not detected without driver, Copilot, what is ACPI\INTC1041\IETM and where can i find the driver ? It proposed this : https://www.catalog.update.microsoft.com/Search.aspx?q=Device+ACPI%5cINTC1041 And it is the correct driver (download the cab, extract it, then update the peripheral, point it to the directory where you extracted the cab). Finally i disconnected one of the two USB M.2, kept only the one with Windows install, on the USB hub with external power. After reboot, in the windows “disk manager”, i finally see my all 5 M.2 Crucial T500 drives :) Next step : raid5 with Storage Spaces :) - control panel > storage spaces > create pool And unfortunatly storage spaces does not work with 4K sector size T5OO drives :( - it expects smaller sector size drives Next thing to explore … Install some WSL linux inside Windows, and try to do some ZFS there … And/or look if someone managed to replace internal power brick by something more powerful (to solve drives drop because not enough power). Or remove 10gbe M.2 and re-try without it.

Finally removed 10gbe M2 NIC (draws too much power for this device), installed Windows by first installing to EMMC then cloning to USB M2 (using DiskGenius), then deleted/remounted/formatted my 5 M2 (crucial T500 4to), then shared to network the individual drives. For now everything looks fine :) Perhaps will retry Truenas later with no 10gbe :)

In fact not so fine, drives continue dropping :

There is existing problem with Physon controllers, some of them solved by bios update :

ME Mini - Truenas - NVME disconnects, USB 5gbe adapter fails

Dres

oh hey Bobo, I have run into the same issue. I sent my first machine back, received a new replacement and I have the same issue.
btw, i have the exact same setup.

« Previous Page