Can you check the output of the following after trying to start it via systemctl?:
journalctl -e -x libvirtd
I’d also see if SELinux is set to Enforcing and if so check the audit logs at /var/log/audit/audit.log to see if it’s triggering something when starting it from the service file.
One typical “can’t start at boot” error for some services is that they start a bit too early and attempt to use network connections that do not exist yet. However, those services do log the errors.
Systemd’s journald should have those, viewable by:
sudo systemctl status libvirtd
sudo journalctl -xeu libvirtd
(Assuming it is the libvirtd.service unit that we are interested in.)
If you do run rsyslogd, then the files it writes to could have a copy, and if the libvirtd explicitly writes to files according to your config, those files should(?) get something.
Logs begin at Wed 2021-12-22 06:32:24 EST, end at Fri 2021-12-24 02:59:01 EST. -
Dec 22 10:01:48 oc003 systemd[1]: Starting Virtualization daemon…
– Subject: Unit libvirtd.service has begun start-up
– Defined-By: systemd
– Support: Red Hat Customer Experience & Engagement - Red Hat Customer Portal
–
– Unit libvirtd.service has begun starting up.
Dec 22 10:01:48 oc003 systemd[1]: Started Virtualization daemon.
– Subject: Unit libvirtd.service has finished start-up
– Defined-By: systemd
– Support: Red Hat Customer Experience & Engagement - Red Hat Customer Portal
–
– Unit libvirtd.service has finished starting up.
–
– The start-up result is done.