do-release-upgrade hangs when systemd-networkd.service fails to restart cleanly; causes sshd issues

[Zachary Zehring]

We upgraded some baremetal machines from bionic to focal. During the d-r-u process, it got stuck when it tried to restart systemd-networkd.service [0]. This also caused the machine to be unreachable through ssh. The only way to work around this issue was to have an existing ssh session on the machine during the upgrade. Then, I was able to restart systemd-networkd.socket and systemd-networkd.service to get things working again; this unstuck the d-r-u process. Additionally, I had to restart sshd for that service to work as well. Other services listening on other ports worked fine. This was a juju unit, and "juju run --machine $num" worked to issue commands remotely on the machine.

[0]
Created symlink /etc/systemd/system/sysinit.target.wants/systemd-pstore.service /lib/systemd/system/systemd-pstore.service.
Job for systemd-networkd.service failed because the control process exited with error code.
See "systemctl status systemd-networkd.service" and "journalctl -xe" for details.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Nick Rosbrook (enr0n)]

Can you please attach the systemd-networkd logs from that time? It would be helpful to see why it failed.

[Launchpad Janitor (janitor)]

[Expired for ubuntu-release-upgrader (Ubuntu) because there has been no activity for 60 days.]