Hi,
I have one ethernet connection called Wired connection 1. This one is started when I boot. Automatically the VPN connection is started as well.
They both use enp2s0 as network device.
Sometimes, I see a notification after a while that a new ethernet connection, called enp2so, has connected, which means the normal one plus the VPN are disconnected.
I then open network manager settings, delete the new enp2s0 connection and connect manually to “Wired connection 1”, which starts the VPN again.
I never had this on other distro’s and I wonder what might cause this. It’s not the end of the world, but when it happens often it is a nuisance.
Is it reproducable in the sense that it happens after each boot, f.e. after VPN connect?
Anything visible through dmesg? journalctl?
Could it be that something is triggered twice, due to some misconfiguration, f.e. VPN started through script and through systemd?
You should start with displaying configured network interfaces
Run the following command when you don’t have a VPN connection, when you have a VPN connection and perhaps again after you believe your machine has re-created the VPN connection.
Reproducable? Well, it did happen a couple of times in the past week, but it’s hard to say when (after being up how long).
I uses network manager as can be seen on the pictures I added. The ethernet connection starts the VPN automatically through the setting in NM.
I will check dmesg and journactl when it happens again to see if I can find something what caused it.
That looks like a timing issue. I’m almost sure a second instance of the Network manager is triggered by VPN not having the eth0 connection yet and trying to start it. But, I could be wrong, just remember a similar issue, some months ago.
How did you set the ‘on boot’ start of the eth0 connection?
How do you invoke the networkmanager? cli ? Applet?
I use the System settings module to setup the network (right-click on the network icon in the system-tray). In there I select:
Automatically connect this network when it is available
All users may connect to this network
Automatically connect to VPN when using this connection
So the ethernet connection is started, a moment later the VPN is up and it can take 10 minutes, maybe half an hour when suddenly the notification comes that I use enp2so connection now instead of the normal Wired connection with the VPN attached to it. It’s not during boot that things go wrong, it’s much later.
I don’t know if your regular network connection is interrupted which would also disconnect your secondary (VPN) connection, but in each case your network services recognizes your machine with its MAC and IPv6 address, it continues to use the same IPv4 address.
On the other hand,
Your VPN interface is re-made each time and is assigned a new IPv4 and IPv6 address.
You should determine whether you’e having basic network connection issues.
That’s because in the router I connected the MAC addresses of all devices to fixed IP addresses so with every boot they have the same address. Still in the computers I use DHCP for easier setup.
Well, it’s not exactly the same cause here the original connection does not return, no matter how long I wait.
@knurpht and @tsu:
I just had the symptom again, almost immediately at start and I did the two codes: dmesg and journalctl -q but apart from lines of text in with the connection is started I see nothing about it being disconnected again and enp2s0 taking over.
Is there any other command I can use to see more info about the network status?
Misunderstanding, the Disconnect and re-Connect is done manually.
If no commands are given the connection stays on enp2s0.
This event has occurred 4 times in the last hour. On one occasion at least the icon showed disconnected although still connected to enp2s0.
nm is set to allow auto negotiate the connection. Its assumed the router is automatically changing parameters and this causes the change of connection.
Although from the PC side nothing appears to have changed with respect to addressing.
the issue was caused by the line in etc/sysconfig/network/config being changed to CHECK_DUPLICATE_IP=“no” (supposedly to speed up booting)
and
the PC ip address (static addressing) being given to another device in the router
hence if the other device was ON, a connection with that address could not be established and then the router re-allocating the connection to a free address
thus the attempt to establish a connection via “vdk-wired” failed but was possible with a different address via “enp2s0”
the cmd ip address though still gave the connection as via the demanded static address and not the new address automatically allocated by the router
resolution
the device addresses allocated in the router were set to different values and the devices themselves set to auto for addressing in NM
now ip address gives
inet 192.168.178.124/24 brd 192.168.178.255 scope global dynamic noprefixroute enp2s0
and the router states
192.168.178.128
if confused by the above, you are not alone
router used: “FRITZ!Box 6490 Cable”
conclusion: user manual needs to be read thoroughly