Proxmox 6.8.12-9-pve kernel has introduced a problem with e1000e Driver and network connection lost after some hours

Zaphood said:
So far, since Thursday no hang/freeze with the 6.14.5-1 kernel on the I219-LM Rev 10... lets hope it stays this way.
Click to expand...
Nice! I just made a one-liner to install my scripts and automatically set them up to simplify the process. If you looked at my current docs on it I'm sure you can see thay it may be a bit complicated for some users haha so I'm simplifying it. I haven't updated my docs yet but the new one-liner approach is this

bash -c "$(curl -fsSL https://raw.githubusercontent.com/TrueBankai416/Scripts/refs/heads/main/Proxmox/install.sh)"

Just copy and paste that whole section starting with "bash" into the terminal
 
Last edited:
Thanks for your work, but i just manually installed the kernel and did a reboot. No fancy parameters or anything else.
If it turns out in a couple of days, that 6.14 did not improve the stability, I will sure use your script then, so thanks in advance ;-)
 
  • Like
Reactions: Twitch6004
Zaphood said:
Being curious: What does your script do different from this community script https://community-scripts.github.io/ProxmoxVE/scripts?id=nic-offloading-fix

?
Click to expand...
I took a quick peek at that script, it should also be permanent i believe. Mine creates a service that actively monitors the kernel logs and network availability. If it sees that it loses internet and that there was a hang in the kernel logs then it will apply the fix. Even if the adapter changes or gets renamed. It will automatically reapply the fix whenever needed
 
Twitch6004 said:
I took a quick peek at that script, it should also be permanent i believe. Mine creates a service that actively monitors the kernel logs and network availability. If it sees that it loses internet and that there was a hang in the kernel logs then it will apply the fix. Even if the adapter changes or gets renamed. It will automatically reapply the fix whenever needed
Click to expand...
Wow, ok, thanks. Way more sophisticated then...

Ok, I am now back to .8-11 Kernel and have applied the community script as a first measure. If that ends up the same way as the test with the .14 Kernel, then I will resort to your script.

And if that doesn't help either, I will quit Proxmox and use Hyper-V, hoping this will solve all my problems... :-D :-D :-D ...or maybe not so....
 
Zaphood said:
Wow, ok, thanks. Way more sophisticated then...

Ok, I am now back to .8-11 Kernel and have applied the community script as a first measure. If that ends up the same way as the test with the .14 Kernel, then I will resort to your script.

And if that doesn't help either, I will quit Proxmox and use Hyper-V, hoping this will solve all my problems... :-D :-D :-D ...or maybe not so....
Click to expand...
My network went back down today but my script successfully fixed it automatically. So seems to be working correctly
 
Last edited:
  • Like
Reactions: Zaphood
phil-schneider said:
Thanks for your script. I also applied it.

I am also having troubles since April. My Home Assistant was not reachable. Unplugging wire helped. I almost thru my router at the wall.... ;-)
Click to expand...
I am currently adding another option to the install script to also optionally include storage scripts that will help analyze and cleanup the filesystem
 
Saidly this morning my whole system was not reachable :-(
Now I am not sure if I installed your script correct.
I just did a fresh proxmox installation with the post install script:
https://community-scripts.github.io/ProxmoxVE/scripts?id=post-pve-install
(which also installs updates)

After that I installed your script on the "main shell", just there.
Then I installed with the community scripts wireguard and homeassistant.
On these two I have NOT installed the script.
Is this correct, or do I also need to install the script on the VM's ?

This morning home assistant was again not reachable.
So I went to the server and tried the "have you unplugged and replugged the wire" trick, no success.
So I went to the server, attacked monitor, logged in and executed some commands:

IMG_20250717_070240.jpg

Then I tried a ping and it was working (did none before).
I used another laptop and connected to proxmos UI, which was working.
The two vm, seems to work, but where not reachable over their IP adress.
Restarted the wireguard vm and it was working again.

So now I am confused.
 
Last edited:
phil-schneider said:
Saidly this morning my whole system was not reachable :-(
Now I am not sure if I installed your script correct.
I just did a fresh proxmox installation with the post install script:
https://community-scripts.github.io/ProxmoxVE/scripts?id=post-pve-install
(which also installs updates)

After that I installed your script on the "main shell", just there.
Then I installed with the community scripts wireguard and homeassistant.
On these two I have NOT installed the script.

This morning home assistant was again not reachable.
So I went to the server and tried the "have you unplugged and replugged the wire" trick, no success.
So I went to the server, attacked monitor, logged in and executed some commands:

View attachment 88110

Then I tried a ping and it was working (did none before).
I used another laptop and connected to proxmos UI, which was working.
The two vm, seems to work, but where not reachable over their IP adress.
Restarted the wireguard vm and it was working again.

So now I am confused.
Click to expand...
From the picture it seems like my scripts worked successfully. I am still actively updating that script so if you run that bash command again you will download the most up to date script which comes with update features and more. you can also refer to this for current manual commands to ensure that the script successfully installs and is monitoring your interface (wondering if the systemd service is working) https://github.com/TrueBankai416/Scripts/blob/main/Proxmox/Readme.md
If you still need help, I'll be available tomorrow
 
Last edited:
thank you very much
But first my questions:
- is it correct to install this just on the main shell?

Here are some observations:
In the picture you can see, that I executed the network-monitor before and this one said: Network ok.
Then I executed the fix script, which seems to fix the proxmox, but not the bridged VM's.

So after everything was working again, I executed your script again just to see whats happening
The scripts printed a red "hardware hand..."
It also printed now a blue" Bridge member summary" with 2 virtual interfaces skipped
But I guess there was none, because everything was working.
Also now the VM's dont work anymore, I have to restart them :-(
 
phil-schneider said:
thank you very much
But first my questions:
- is it correct to install this just on the main shell?

Here are some observations:
In the picture you can see, that I executed the network-monitor before and this one said: Network ok.
Then I executed the fix script, which seems to fix the proxmox, but not the bridged VM's.

So after everything was working again, I executed your script again just to see whats happening
The scripts printed a red "hardware hand..."
It also printed now a blue" Bridge member summary" with 2 virtual interfaces skipped
But I guess there was none, because everything was working.
Also now the VM's dont work anymore, I have to restart them :-(
Click to expand...
My situation may differ from yours which may cause my script to need improvements. Yes the main shell is fine but I created a folder called "tools" and run it in there. If I remember correctly the script scans the logs for hangs within the last 10min. I'll have to double check that tomorrow. As for as the VMs, I use mostly CTs but I do run a vm with haos. Which is still working for me. So I would need more info on your vm setup to compare to my own
 
Last edited:
Twitch6004 said:
My situation may differ from yours which may cause my script to need improvements. Yes the main shell is fine but I created a folder called "tools" and run it in there. If I remember correctly the script scans the logs for hangs within the last 10min. I'll have to double check that tomorrow. As for as the VMs, I use mostly CTs but I do run a vm with haos. Which is still working for me. So I would need more info on your vm setup to compare to my own
Click to expand...
My setup is pretty simple:
HomeAssistant with this:
https://community-scripts.github.io/ProxmoxVE/scripts?id=haos-vm

Wireguard with this:
https://community-scripts.github.io/ProxmoxVE/scripts?id=wireguard
 
Twitch6004 said:
I'll have to take a look at these tomorrow, so your issue is that proxmox works but you cannot access the VMs? Is this an old issue or did it happen after using my script?
Click to expand...
Not sure, sorry.

In the morning, HA was not reachable. (also wireguard was reachable)
I did not check the proxmox UI.

So I used console to run your scripts.
After that I connected to proxox UI and saw that the wireguard and HA was running. But both were not reachable. So I restarted them.


So what also could have happend:
- in the night was again a *** hardware hang
- your script might have restarted proxox nic interface (i dont know)
- but the VM (HA) and LMX (wireguard) somehow did not "restart" the network or did not updated
 
Last edited:
You must log in or register to reply here.
Top Bottom