Comment 8 for bug 1863873

I can confirm this behavior after update to systemd 237-3ubuntu10.39 on privileged and unprivileged containers. We classify this bug as critical because in the next monthly update cycle via ansible orchestration we would expect failing all Bionic LXC containers. The only workaround seems to manually define a static network configuration in /etc/netplan/10-lxc.yaml via lxc-attach. But this is not an acceptable solution for 180 servers we run.

Some our Plesk servers already failed completely due to automatic upgrades Plesk triggered daily. Websites were down for hours until we figured out the reason.

I would appreciate that someone solves this problem due to urgency.

Thank you in advance.