4

Problem

Almost all networking fails. Domain names will not resolve, websites won't load, and even most ping commands on a local network will fail. For example:

  • ping localhost results in 100% packet loss and will always time out if a threshold is set.
  • ping 8.8.8.8 results in 100% packet loss and will always time out if a threshold is set.
  • ping ::1 (localhost on IPv6) works.
  • ping 2001:4860:4860::8888 (Google DNS on IPv6) results in ping: connect: Network is unreachable.

The outputs of ip addr indicate that my ethernet adapter is not found among the devices listed, and lshw also shows it as missing.

I assume from this that there are some remaining uninstalled drivers or that the upgrade did not successfully upgrade the device drivers, in addition to potential other networking issues generated by this upgrade process. So there may be more than one problem here; the ping issue is the most fundamental, as it prevents me from reinstalling packages that might fix additional issues.

I have attempted some fixes to manually restore networking capabilities, such as adding an entry for my wifi device in /etc/network/interfaces and also again in /etc/netplan/01-network-manager-all.yaml. These were not successful. I don't know enough to try more fixes, which is why I am asking here.

Additional Context

systemctl status networking

There may be potential issues with resolved.

$ systemctl status networking
    ifup[1634]: /etc/network/if-up.d/resolved: 12: mystatedir: not found

systemd-resolved

The systemd / systemctl logs indicate a degraded state for systemd-resolved - it loops back and forth between two degraded states as indicated by the logs.

$ systemctl status systemd-resolved:
    Using degraded feature set UDP instead of TCP for DNS server 192.168.1.1
    Using degraded feature set TCP instead of UDP for DNS server 192.168.1.1
    ...

I have saved the output of the wireless-info shell script which has command outputs with details redacted; it is available as a GitHub gist for maximal information.

afoltzm
  • 41
  • 1
  • 3

2 Answers2

1
    ifup[1634]: /etc/network/if-up.d/resolved: 12: mystatedir: not found

There are lines 11-13 in /etc/network/if-up.d/resolved, it looks like:

if systemctl is-enabled systemd-resolved > /dev/null 2>&1; then
    mystatedir statedir ifindex interface

And later in line 26:

    mystatedir=/run/network

This means: word with variable name mystatedir is used as executable file's name placed at a $PATH. This script has not been tested. Most probably smth. like local keyword missed at line start.

May be all the rest has not been verified too.

Asc
  • 103
0

If anyone else comes to this after upgrading to 22.04 on Digitalocean, or potentially other cloud providers, the following might help.

I was unable to access my droplet after upgrading from 20.04 to 22.04 on Digitalocean, and from the recovery console KVM, was able to see no networking available with the classic "Network is unreachable" from a direct IP ping.

ip a showed no ip address or gateway given.

Turns out, in /etc/network/interfaces.d/50-cloud-init.cfg there were a few new lines added:

# control-alias eth0
iface eth0 inet static
    address 10.17.0.6
    netmask 255.255.0.0

These overrode my eth0 interface configuration with the correct ip address and gateway, and caused the interface to have the incorrect configuration.

Removing them resolved the issue.

xxx
  • 105