One of our CI jobs seemingly breaks systemd on our hosts resulting in the following error when interacting with it (e.g. trying to reboot):
System has not been booted with systemd as init system (PID 1). Can't operate.
Failed to connect to bus: Host is down
Failed to talk to init daemon.
Clearly this is a lie as ps confirms.
"Host is down" solutions recommend usually to issue some systemctl command but that just caused the same behavior.
However, you can issue those commands by sending signals as the manpage reveals. In our case sending SIGTERM to PID 1 solved the issue (though not sure what causes the issue in the first place :/).
Upon receiving this signal the systemd system manager serializes its state, reexecutes itself and deserializes
the saved state again. This is mostly equivalent to systemctl daemon-reexec.