Ethernet connection drops at random times

Hi.

My Endeavouros just drops the ethernet connection at random times. Sometimes the connection just drops when I’m afk for 5 minutes, sometimes the connection drops when I’m using the computer.

I haven’t had this issue on other distributions. I am running endeavouros on a pc and not on a laptop. On wifi I haven’t had this issue on the same computer. There is a laptop connected to wifi/ethernet on the same network and it doesn’t have this issue.

Any ideas how to fix it?

Here is what I have tried to fix it:

  1. update system
  2. tried different network cables (currently on a completely new cable and the issue still presists)
  3. logging back out and in or restart fixes the issue (but it’s very counterproductive so I’d like to find a better fix)

I will happily try different fixes and provide you with any logs that are needed to figure this out. If logs or info is needed, please tell me what command I have to use to get the log (I’m a beginner).

Thank you :slight_smile:

Can you post your hardware. Post the url

inxi -Faz | eos-sendlog

1 Like

Thank you! Here is the URL!

https://0x0.st/XKJQ.txt

Sometimes it’s one of two things:

  1. flimsy cable or poor connections
  2. if it’s an external USB card maybe it overheats, or draws insufficient power.

It might be worth updating the UEFI Firmware (Bios). You have the original version 1003. There have been 15 more Bios updates since then. It may or may not help with this issue but worth trying.

1 Like

I knew you’d say this! :wink:
And it is a good advice.

1 Like

Hey. Thank you for the replies everyone. I really appreciate your help :slight_smile:

I updated the bios to latest firmware and still the same issue presists (connection loss at random times).

The cable is 100% fine, the connections are 100% fine, endeavouros is the only linux distro where I have this issue.

There is no external usb card that draws insufficient power or overheats. The electric system in my home was installed in 2024 (its new and works), the ethernet cable goes directly into the ethernet slot in motherboard (no external usb).

Any more ideas that may fix this issue?

Are you dual booting with Windows?

You could compare the inxi output above with the output from other distros, there might be some clue.

Hey :slight_smile: No. I’m not dual booting with windows. One of the harddrives has endeavour, one has another linux distro, one is just for files.

What does this show?

sudo dmesg | grep e1000e

Hey. I really appreciate your help! :))

So I tried " sudo dmesg | grep e1000e"
Tried it after internet went away.
Tried it when internet was still here

  1. (internet had went away when I tried this one)

[ 4.000283] e1000e: Intel(R) PRO/1000 Network Driver
[ 4.000285] e1000e: Copyright(c) 1999 - 2015 Intel Corporation.
[ 4.000572] e1000e 0000:00:1f.6: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[ 4.244571] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): registered PHC clock
[ 4.312621] e1000e 0000:00:1f.6 eth0: (PCI Express:2.5GT/s:Width x1) 2c:fd:a1:6e:f7:01
[ 4.312624] e1000e 0000:00:1f.6 eth0: Intel(R) PRO/1000 Network Connection
[ 4.312676] e1000e 0000:00:1f.6 eth0: MAC: 12, PHY: 12, PBA No: FFFFFF-0FF
[ 4.471745] e1000e 0000:00:1f.6 enp0s31f6: renamed from eth0
[ 9.679116] e1000e 0000:00:1f.6 enp0s31f6: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

  1. (had internet)

[ 4.000283] e1000e: Intel(R) PRO/1000 Network Driver
[ 4.000285] e1000e: Copyright(c) 1999 - 2015 Intel Corporation.
[ 4.000572] e1000e 0000:00:1f.6: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[ 4.244571] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): registered PHC clock
[ 4.312621] e1000e 0000:00:1f.6 eth0: (PCI Express:2.5GT/s:Width x1) 2c:fd:a1:6e:f7:01
[ 4.312624] e1000e 0000:00:1f.6 eth0: Intel(R) PRO/1000 Network Connection
[ 4.312676] e1000e 0000:00:1f.6 eth0: MAC: 12, PHY: 12, PBA No: FFFFFF-0FF
[ 4.471745] e1000e 0000:00:1f.6 enp0s31f6: renamed from eth0
[ 9.679116] e1000e 0000:00:1f.6 enp0s31f6: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

Hmm … not seeing any difference.

vs

:man_shrugging:

Sorry, have to strongly disagree here. Terrible advice, and @ricklinux is giving it like candy to everyone lol.

  • In 99.9% of the cases it will not resolve the problem at hand.
  • Greatly increases the risk of bricking something, especially with novice users
  • In most cases you will need wind*ws to do the update
  • If the BIOS was not updated until now, there must be some strong reason for it, either user did not care, or they were not aware of it
  • In most of the cases where this was advised the system worked fine until some update or configuration change happened
1 Like

What is terrible advice?

:candy:

Edit: Updating UEFI firmware (Bios) is one of the most important updates when it comes to motherboards. It may not fix a particular issue such as this but is very important for all other aspects of the hardware.

2 Likes

One thing worth mentioning is that updating bios should be done sequentially: all the intermediate versions should be flashed in order.

Again strongly disagree with this, the user would have updated already if something was malfunctioning badly. If it did not happen until the point the point the question was asked, there is no need to update.

Don’t get me wrong, I agree that BIOS/UEFI should be updated in general, but it usually distracts the user from the issue at hand, and greatly increases the risk of bricking the hardware.

I’d offer that the onboard ethernet controller failing inexplicably, is one possible scenario.

I do agree that when it comes to firmware updates, as there are risks, some understanding of what the update might offer is important. In this case, the motherboard was on version 1003 (Jan 2018), and updated to 2701 (Aug 2021), spanning something like 15 update releases.

Multiples of those updates included numerous ethernet related updates(ME firmware), and vaguer fixes to “system stability”. It doesn’t seem unreasonable that any one of those might have resolved the issue. Unfortunately it didn’t, but that’s an easy step that’s at least ruled out now.

@quxup, I have a few thoughts:

  • Have you modified MTU settings, or enabled jumbo frames on any part of your network?
  • Are you using a VPN either on this PC, or your router? (can also impact MTU)
  • Can you confirm that when the connection drops, your LAN is not accessible? (try simultaneously connecting to your router’s web interface, and Google for example)

At the time you experience a dropout, can you share the output of:
$ ifconfig

I don’t agree that this is necessary unless the manufacturer specifically specifies that a certain version has to be upgraded before another. This does happen at times.

1 Like