SER8's LAN stops working when I put a second SSD in the slot above the WiFi card

jaymebc

I bought a SER8 AMD 8845HS with 32GB of RAM and a 1TB SSD. I bought a Kingston 1TB SSD, NVMe M.2 2280 PCIe 4.0 to use in the second SSD slot. This slot is above the WiFi card.

When I install it in this slot, the LAN stops working. I removed it and it started working again. I switched the SSDs (the factory one above the WiFi card) and it stopped working. I even tried to use both SSDs after removing the WiFi card, also without success.

I couldn’t find a solution on the forum. I’ve already tried resetting the CMOS (I held it for 10 seconds and waited 15 minutes to test) without success.

To summarize: when I try to use an SSD in the slot above the WiFi card, the LAN stops working.

Has anyone experienced this? Suggestions?

Thanks in advance!

support1

jaymebc Hi there, we will test it and get back to you later.

support1

jaymebc Hi there,

We did three tests, the SSD works fine.
Here are the test video :
https://mega.nz/file/fy52CS7T#xxHvRah8R_WXJR62-uBQU4OPkkl496z_DWqHq3JCKOY
https://mega.nz/file/eqBwkLbb#rZbSbdDsSZ5OjQlIwrs9FbgJjXIvlpg1otNjIh93KPc
https://mega.nz/file/irADURbL#RDfu9fJuhjIwv_4vayxZm8RdV-cEXNSGgYM5H3ep3rE

① Try a different NVMe SSD in the second slot to rule out a faulty Kingston drive or compatibility quirks.
② When the SSD is installed, check if the LAN controller disappears or shows errors (e.g., Code 12: “Not enough resources”). This confirms a resource conflict.
③Please press delete key as soon as you turn on the PC, so that you can go to BIOS. Please send us a picture of the Main page.
We will check if you need to update the BIOS.
④Please Update LAN drivers and chipset drivers: https://url.bee-link.cn/xv1s

jaymebc

Thanks for the quick response!

I tested it here at home with two different NVMe SSDs in the second slot and the result was the same. However, I think the problem is not in the hardware, but in the software.

As soon as I received the computer, I turned it on and verified that everything was working fine. Since the goal is to use it as a platform to run virtual machines on my home network, I installed Proxmox without any difficulty. Except that the second slot does not work with Proxmox (the second slot was only occupied after the installation of Proxmox). So, I think it must be a driver problem.

When I received your message, I reinstalled Windows and everything worked perfectly.

So the question remains: could it just be a lack of drivers? If so, do you have the driver for Proxmox? I imagine that drivers for Debian/Ubuntu should work, right?

support1

jaymebc
Hi there,

Check for PCIe Lane/Resource Conflicts (Likely Cause)
The SER8’s second M.2 slot (above the WiFi card) shares PCIe lanes or IRQ resources with the LAN controller. When the SSD is detected, the system may not allocate resources correctly in Proxmox (Linux-based), whereas Windows handles it automatically.

Steps to Diagnose:
In Proxmox, run these commands to check hardware conflicts:

bash
lspci -vv # Check if LAN controller or SSD shows errors (like “no resources”)
dmesg | grep -i “error|pcie|resource” # Look for allocation failures
If the LAN controller disappears or shows errors (e.g., Code 12), it’s a resource conflict.

Workarounds:
Update the BIOS: Please press delete key as soon as you turn on the PC, so that you can go to BIOS. Please send us a picture of the Main page.We will check if you need to update the BIOS.

Proxmox/Linux Driver Issues
Proxmox uses a Linux kernel, which may lack proper drivers for:

The Realtek LAN controller (common in mini-PCs).

The NVMe SSD controller (if it’s a newer model).

Solutions:
Install Latest Drivers:

LAN Driver: Realtek often requires manual driver installation in Linux. For Debian/Proxmox:

bash
apt update
apt install firmware-realtek r8168-dkms
NVMe Driver: Most modern kernels support NVMe, but if the SSD is very new, update the kernel:

bash
apt install pve-kernel-6.2 # Install a newer Proxmox kernel
Blacklist Conflicting Drivers:
If the kernel loads the wrong driver (e.g., r8169 instead of r8168), blacklist it:

bash
echo “blacklist r8169” >> /etc/modprobe.d/blacklist.conf
update-initramfs -u
reboot

Test with a Live Linux Distro
Boot a Ubuntu Live USB to see if the issue persists outside Proxmox:

If Ubuntu detects both SSDs and LAN, the problem is Proxmox-specific (driver/config).

If Ubuntu also fails, it’s likely a BIOS/hardware issue.

jaymebc

Hello,

I am attaching a photo of the BIOS.

BIOS -> https://mega.nz/file/hc5h3ITL#k5FYk5J7gkZvyZx5cnk2scDQ-2oMCrLxaHvSB5uZQhg

Something interesting happened. When I first installed Proxmox, I installed it in the first M.2 slot (on the original SSD) and only then did I add the second SSD and had the problem I reported to you.

This time, to perform the tests you requested, I left both SSDs in the PC and installed Proxmox on the SSD in the second M.2 slot. The installation went without any problems and this time the LAN port worked perfectly!

https://mega.nz/file/RdhFjKKQ#cYcGNlWKkWoUhrsr4V_Bcc20c6PjlNGOJdG8oPnhbuU
https://mega.nz/file/YI5hkSwK#DvWpoz0BeWTomkUK1y594sYFYoBQiUoejIfLiOgVkxk

About the commands you requested:

lspci -vv -> attached, I found no errors
https://mega.nz/file/RRgQQLjT#OSPsUw411vCEVmB-sAkIZdw2e-_9BbtC5OlgSiB_33c

The second command gives error:
root@pve:~# dmesg | grep -i “error|pcie|resource”
-bash: pcie: command not found
-bash: resource”: command not found

“apt install firmware-realtek r8168-dkms” also gives an error:
root@pve:~# apt install firmware-realtek r8168-dkms
Reading package lists… Done
Building dependency tree… Done
Reading state information… Done
Package firmware-realtek is not available, but is referred to by another package.
This may mean that the package is missing, has been obsolete, or is only available from another source
However the following packages replace it: pve-firmware
E: Package ‘firmware-realtek’ has no installation candidate
E: Unable to locate package r8168-dkms

The Proxmox version is the most current (8.4 - Kernel 6.8.12-9-pve). Thus, as lspci shows that r8169 is already the default Linux kernel driver for this controller and r8168 would be outdated. Is this correct? From what I’ve read, the r8168-dkms package provides an alternative driver for Realtek controllers developed by Realtek itself instead of the generic r8169 driver maintained by the Linux community. r8168 seems to offer better performance or stability in specific scenarios (like 2.5GbE networks under heavy load) but it would not be included in the standard Proxmox VE repositories (hence the error). I don’t know if r8168-dkms is compatible with the Proxmox kernel without tweaks (and I wouldn’t know how to do that).

I didn’t update the kernel because it is already at the latest version.

The command “dmesg | grep -i r8169” returned the following:
root@pve:~# dmesg | grep -i r8169
[ 0.800164] r8169 0000:02:00.0 eth0:RTL8125B, 70:70:fc:07:f0:59, XID 641, IRQ 52
[ 0.800168] r8169 0000:02:00.0 eth0: jumbo features [frames: 9194 bytes, tx checksumming: ko]
[0.863723] r8169 0000:02:00.0 enp2s0: renamed from eth0
[ 5.056252] r8169 0000:02:00.0 enp2s0: entered allmulticast mode
[5.056310] r8169 0000:02:00.0 enp2s0: entered promiscuous mode
[ 5.080439] RTL8226B_RTL8221B 2.5Gbps PHY r8169-0-200:00: attached PHY driver (mii_bus:phy_addr=r8169-0-200:00, irq=MAC)
[5.245239] r8169 0000:02:00.0 enp2s0: Link is Down
[ 7.531248] r8169 0000:02:00.0 enp2s0: Link is Up - 1Gbps/Full - flow control off
[496.853754] r8169 0000:02:00.0: invalid VPD tag 0×00 (size 0) at offset 0; assume missing optional EEPROM
root@pve:~#

From what I understand, no errors, right? (except for the optional EEPROM being missing but I understand that this is a warning and not an error, and does not affect the functioning of the LAN).

I tried resetting the machine a few times and everything works normally.

Do you suggest anything else to be done?

support1

Hi there,

① Add a kernel parameter to force reallocation of IRQ resources:
Edit the Grub configuration:
bash
nano /etc/default/grub

Append pci=realloc to the GRUB_CMDLINE_LINUX line:
bash
GRUB_CMDLINE_LINUX=“pci=realloc”

Update Grub and reboot:
bash
update-grub
reboot

This forces the kernel to reallocate PCIe resources at boot, which may resolve conflicts.

②Since installing Proxmox on the upper slot (second SSD) works, you can use this as a workaround:
Clone your original SSD to the upper slot (e.g., via dd or a GUI tool like GParted).