Network Issues on Proxmox - NETDEV WATCHDOG transmit queue timed out (e1000e)

jojd

New Member
Oct 2, 2024
1
0
1
Hello everyone,

I’m experiencing a network issue on my Proxmox server and could use some help troubleshooting. The server has been running smoothly for a while, but recently it started losing network connectivity. It works fine for a few minutes after rebooting the router (Unifi router), but then it disconnects again. I've attached a screenshot of the console showing the repeated error messages.

Error Message:

NETDEV WATCHDOG: enp0s31f6 (e1000e): transmit queue 0 timed out

System Details:

  • Proxmox Version: [8.2.4]
  • Router: Unifi router, static IP
 

Attachments

  • image.jpg
    image.jpg
    286 KB · Views: 23
Same here too. Is there a fix?
I do have one Win 11 VM on the host if that helps narrow it down.
 
Hi,

Has anyone managed to resolve this?

I am new to Proxmox and love it. Running a Win11 VM, an Ubuntu VM running docker and portainer with containers for Pihole, MQTT. In addition I have a HomeAssistant VM.

When running, everything runs beautifully, but I am getting the same NETDev watchdog errors, and it is bringing the Proxmox server down completely which then takes out everything.

Extra problem I have is I work away at the moment and when it breaks my partner and family are stuffed :-)

Thanks

Andy
 
Post the contents of your /etc/network/interfaces file here, and I'll show you how to fix this issue. I finally fixed mine. You likely have an Intel network interface and there is a known issue with the driver.
 
Thanks avpman, for taking the time to respond to this. I genuinely have tried to sort this myself. I Googled and came across this (pre-up and post-up lines):

Interfaces config below:

auto lo
iface lo inet loopback

iface enp0s31f6 inet manual

auto vmbr0
iface vmbr0 inet manual
bridge-ports enp0s31f6
bridge-stp off
bridge-fd 0
bridge-vlan-aware yes
bridge-vids 2-40
pre-up ethtool -K enp10s0f0np0 tso off #<- pre disables tso on the physical nic
post-up ethtool -K vmbr0 tso off #<- post disables tso on the virtual bridge


auto vmbr0.10
iface vmbr0.10 inet static
address 192.168.10.254/24
gateway 192.168.10.1

iface wlp0s20f3 inet manual

source /etc/network/interfaces.d/*

I disabled TSO and that seemed to fix it.....until a reboot.


Once again thanks for looking at this...i want to learn.

Andy
 
Andy,

You're very close! You need to specify the full path to 'ethtool' in the interfaces file. Here's mine. Be sure to verify your path to ethtool on your system and you only need the post-up entry.

post-up /usr/sbin/ethtool -K enp0s31f6 gso off gro off tso off

Please post back with your results. I sure hope this helps you. I pulled my hair out for weeks over this.

Jim
 
Update- Stupidly had the wrong interface name in the config. Changed and after a reboot TSO is off.
 
Andy,

You're very close! You need to specify the full path to 'ethtool' in the interfaces file. Here's mine. Be sure to verify your path to ethtool on your system and you only need the post-up entry.

post-up /usr/sbin/ethtool -K enp0s31f6 gso off gro off tso off

Please post back with your results. I sure hope this helps you. I pulled my hair out for weeks over this.

Jim
ahhh, posted the update before seeing this. Will add the path in now. Thank you!
 
Andy,
You're very welcome. I am happy I was able to give back. This plagued me for weeks until I cobbled together the fix from soooooo many posts from people having this exact issue.
Jim
 
I was facing the same issue and, as suggested by @avpman, I’ve updated my network configuration. Hopefully, this will resolve the problem. I have a small question:

I’ve been running Proxmox on my Intel NUC for the past two years without any issues. I’ve read that the problem is due to a driver issue for Intel adapters, which is definitely my case.

So, I’m wondering if the driver issue was introduced in one of the recent kernel updates or if it’s an old issue that’s now manifesting due to a Proxmox update.

Thanks.
 
Last edited: