Problem with Intel I219-V Ethernet Adapter and e1000e driver

M

marmour

New Member
Sep 25, 2025
6
1
3
I am having a problem with my PROXMOX PVE not upping or showing the Intel I219-V onboard ethernet with the e1000e driver. This is a MSI CODEX R2 system with an onboard ethernet. This is the only available ethernet except for a WIFI 6 adapter. It has been very difficult to get a network connection to try to update the packages. I did manage to get a USB-C to ethernet adapter working to get a network connection. Package updates didnt help the situation. The error from the DMESG output is NVM Checksum Not Valid. Error Code -5. This is a new install on Promox 9.1 so there is no way to fall back to an older version. I have tried many things as outlined in other posts such as:
1. apt update, apt full-update, apt dist-update
2. Disable ASPM in BIOS
3. Add pcie_aspm=off in GRUB. And do update-grub
4. Since the driver isnt loaded it doesnt show in the ip a
5. lspci -nnk | grep -iA3 net does show the I219-V

pve-manager 9.1.1 kernel 6.17.2-1-pve

I have read that updating the driver for e1000e and have it ignore the checksum fail is a possible solution but it seems complicated and I havent found a straight forward post to walk me through it. Plus since I am not a Linux expert things like download driver, use ethtool, update driver etc are not clear as to where to go and how to get these items and get them in the Proxmox PVE environment to do the updates.

I am looking for help in this please. Is there a way to get this working without manually patching the driver. Is there somewhere a fixed driver can be found to get the update on to the Proxmox server. I would prefer not to run on a down level kernel as of course fixes, features etc would not be available. And of course the suggestion to fix any error is to update to the current version.
 
Don’t have that hardware, but maybe this helps:
M

[SOLVED] Post in thread 'Intel NIC e1000e hardware unit hang'

Just came here to say that the following fixed my e1000e intel nic hanging issue. Both of the following have successfully worked around this stupid issue:

* Switch my heavy usage VM's to use the emulated e1000e NIC at the VM level, instead of virtio (limited testing, but worked for about 24 hours, where previous crashes were every ~hour). Choose your VM > Hardware > Network Devices > Go to "Model" drop down > Select e1000e
-OR-
* disable tcp-segmentation-offload on the physical intel NIC's (I did not do the vmbr interface): (This is what I stuck with, not the VM driver change above)...
 
SteveITS said:
Don’t have that hardware, but maybe this helps:
M

[SOLVED] Post in thread 'Intel NIC e1000e hardware unit hang'

Just came here to say that the following fixed my e1000e intel nic hanging issue. Both of the following have successfully worked around this stupid issue:

* Switch my heavy usage VM's to use the emulated e1000e NIC at the VM level, instead of virtio (limited testing, but worked for about 24 hours, where previous crashes were every ~hour). Choose your VM > Hardware > Network Devices > Go to "Model" drop down > Select e1000e
-OR-
* disable tcp-segmentation-offload on the physical intel NIC's (I did not do the vmbr interface): (This is what I stuck with, not the VM driver change above)...
Click to expand...
Sorry this is not the same issue. This one is a e1000e hang of the nic. The nic was working then hung. My issue is the nic isnt even showing due to the fact the driver load had an error with the Checksum Not Valid Error Code -5
 
This is a different issue from the hardware unit hang — the "NVM Checksum Not Valid" error means the NIC's EEPROM contents are corrupted or were never properly flashed at the factory. The e1000e driver refuses to load because it cannot trust the hardware configuration data.

A couple of options. First, check if MSI has a BIOS update for the CODEX R2 — some board revisions shipped with incorrect NIC EEPROM data and the BIOS update reflashes it. Second, if that does not fix it, Intel provides the eeupdate tool (sometimes called lanconf or bootutil depending on the generation) which can reflash the NIC EEPROM from a DOS or Linux boot environment. You would download the Intel Ethernet Connections Boot Utility from Intel, boot from USB, and run eeupdate to write a clean NIC firmware image.

Third, as a temporary workaround on Linux you can try loading the driver with the checksum check bypassed, but this is not recommended for production. The USB-C adapter is honestly fine as a permanent solution too if the NIC EEPROM cannot be recovered.
 
PaddraighNet said:
This is a different issue from the hardware unit hang — the "NVM Checksum Not Valid" error means the NIC's EEPROM contents are corrupted or were never properly flashed at the factory. The e1000e driver refuses to load because it cannot trust the hardware configuration data.

A couple of options. First, check if MSI has a BIOS update for the CODEX R2 — some board revisions shipped with incorrect NIC EEPROM data and the BIOS update reflashes it. Second, if that does not fix it, Intel provides the eeupdate tool (sometimes called lanconf or bootutil depending on the generation) which can reflash the NIC EEPROM from a DOS or Linux boot environment. You would download the Intel Ethernet Connections Boot Utility from Intel, boot from USB, and run eeupdate to write a clean NIC firmware image.

Third, as a temporary workaround on Linux you can try loading the driver with the checksum check bypassed, but this is not recommended for production. The USB-C adapter is honestly fine as a permanent solution too if the NIC EEPROM cannot be recovered.
Click to expand...
Hi thanks, for the response. Yes, I figured that was the problem. I checked and the BIOS version I have is the same as the one available on the MSI site. However, if not flashed properly do you think if I try to update the BIOS anyway that would reflash the NIC EPROM?
Secondly, yes I thought about trying to reflash the NIC EEPROM however I was unaware on how to do that. I had heard about reflashing it. Is there a post somewhere to indicate where to get the bootutil, eeupdate tool, lanconf. Or if you can provide the URL and steps.
Thirdly, how can you load the driver by bypassing the checksum. Is there a thread or you could provide the steps to do this.
I would prefer not to continue to use the USB-C only because I have a onboard proper NIC to use and the USB-C NIC is really a workaround. I am not sure if the throughput of USB-C vs Onboard is slower.
Your help would be appreciated. At least these are constructive recommendations.
 
Last edited:
marmour said:
Hi thanks, for the response. Yes, I figured that was the problem. I checked and the BIOS version I have is the same as the one available on the MSI site. However, if not flashed properly do you think if I try to update the BIOS anyway that would reflash the NIC EPROM?
Secondly, yes I thought about trying to reflash the NIC EEPROM however I was unaware on how to do that. I had heard about reflashing it. Is there a post somewhere to indicate where to get the bootutil, eeupdate tool, lanconf. Or if you can provide the URL and steps.
Thirdly, how can you load the driver by bypassing the checksum. Is there a thread or you could provide the steps to do this.
I would prefer not to continue to use the USB-C only because I have a onboard proper NIC to use and the USB-C NIC is really a workaround. I am not sure if the throughput of USB-C vs Onboard is slower.
Your help would be appreciated. At least these are constructive recommendations.
Click to expand...
I have since flashed the BIOS again. No change. I tried to do the NVAM update using the tool in Windows, nvmupdatew64e it indicated that the I219-V was not a valid device. This happens because it is on the M/B and not PCI-E so the tool wont do it. I would be interested in having the driver bypass the checksum so if you could providing either steps or a link on how to go about that. This is not a production system so it wouldnt be an issue.
 
You must log in or register to reply here.
Top Bottom
Back