Ethernet connects to "lo" interface and doesn't uses the manual IP configuration

ricen · April 16, 2023, 7:34pm

Hello,
I’m on Debian Sid with Plasma 5.27.2 and I have a problem with the ethernet configuration.

If I connect the ethernet cable, i see that te “lo” interface is connected and the Wired connection is not. This is a problem because I can’t set the manual IP/DNS parameters.

Eg.
I’ve set the Wired Connection to use the 192.168.1.XXX manual IP. Despite this configuration, on the router I see the PC on a DHCP-assigned IP like 192.168.1.Y

I don’t know if it’s normal that the “connected” network is “lo”… I thought to change the “lo” settings to use the 192.168.1.XXX IP, but I think I’ll break everything else

Any idea on how to fix this problem so I can use my manual configuration? It always worked on any distro and with previous versions of Plasma…

richarson · April 17, 2023, 2:46am

‘lo’ is the loopback interface and it should have IP 1270.0.1 or similar, I sugest you don’t change it or indeed you’ll break some things

NetworkManager should ignore ‘lo’ but there was a bug in a recent version which made it visible.
If it bothers you, you can add this to /etc/NetworkManager/NetworkManager.conf:

[keyfile]
unmanaged-devices=interface-name:lo

If you already have a section named keyfile with a unmanaged-devices entry, simply add ‘lo’ to it, e.g.:

[keyfile]
unmanaged-devices=interface-name:docker0,interface-name:lo

richarson · April 17, 2023, 2:50am

In your wired configuration, make sure to have “Automatically connect…” checked.

I’m not sure about the names in english, here’s a screenshot that I hope will help:

ricen · April 17, 2023, 5:23am

Thanks for the reply.

I did some test but I can’t make it work. Automatic connection is enabled, but I noticed the connection priority was set to “-100” and you have “0”. So I’ve set “0” too but it still doesen’t work I’ve also tried to set it to “50” just for testing.

Another strange thing I noticed: the Wired connection was “last used” on March 14th… I tried to delete it and create a new one, but it still doesn’t work. It doesn’t connect at all because Plasma settings say that the new wired last connection is “never”

richarson · April 17, 2023, 3:10pm

Weird, maybe your ethernet port is not detecting the cable connection somehow?

Can you post the output of the command /sbin/ip address show before and after connecting the cable?

ricen · April 17, 2023, 4:38pm

You can find below the output, thanks again for your help.

FYI this PC (Intel NUC 12th gen) has another strange behavor related to the ethernet connection and maybe to this problem too:

If I boot the PC while the cable is NOT connected, the ethernet connection doesn’t work if I connect the cable. I have to connect the cable and reboot the system.
If I boot the PC while the cable IS connected, I can connect/disconnect the cable without any problem and it works.

So I can suppose that my problem goes beyond the static IP configuration

NOT connected:

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp100s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN group default qlen 1000
    link/ether 48:21:0b:3d:ce:ef brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.10/24 brd 192.168.1.255 scope global dynamic enp100s0
       valid_lft 21446sec preferred_lft 21446sec
    inet6 fe80::4a21:bff:fe3d:ceef/64 scope link 
       valid_lft forever preferred_lft forever
3: wlo1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether ea:4c:35:9e:93:6c brd ff:ff:ff:ff:ff:ff permaddr a0:59:50:7f:f6:93
    altname wlp0s20f3

CONNECTED TO THE CABLE:

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp100s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 48:21:0b:3d:ce:ef brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.10/24 brd 192.168.1.255 scope global dynamic enp100s0
       valid_lft 21398sec preferred_lft 21398sec
    inet6 fe80::4a21:bff:fe3d:ceef/64 scope link 
       valid_lft forever preferred_lft forever
3: wlo1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether ea:4c:35:9e:93:6c brd ff:ff:ff:ff:ff:ff permaddr a0:59:50:7f:f6:93

richarson · April 17, 2023, 11:45pm

OK, assuming that enp100s0 is your wired connection it gets an IP 192.168.1.10.
I also assume that’s not the fixed IP you want but one provided by DHCP.

Your comment about the connection issues when you boot your NUC without the cable makes me suspect of a driver/kernel problem.

I had a similar problem at work with a particular motherboard and CentOS 7 that when the system booted with the cable it worked but if I disconnected and reconnected the cable, the network wouldn’t come up again or would be flaky.

With Alma 8 no such problems.

You might want to try asking on a Debian forum to see if a different kernel/driver version might help, or if someone else has the same problem.

ricen · April 18, 2023, 2:22am

yes, I should probably try to fix that problem first. Thanks for your help