Important: The openrc libvirtd init script is now broken up into two
separate services: libvirtd, that solely handles the daemon, and
libvirt-guests, that takes care of clients during shutdown/restart of the
host. In order to reenable client handling, edit /etc/conf.d/libvirt-guests
and enable the service and start it:

	$ rc-update add libvirt-guests
	$ rc-service libvirt-guests start


For the basic networking support (bridged and routed networks) you don't
need any extra software. For more complex network modes including but not
limited to NATed network, you can enable the 'virt-network' USE flag. It
will pull in required runtime dependencies


If you are using dnsmasq on your system, you will have to configure
/etc/dnsmasq.conf to enable the following settings:

	bind-interfaces
	interface or except-interface

Otherwise you might have issues with your existing DNS server.


For openrc users:

	Please use /etc/conf.d/libvirtd to control the '--listen' parameter for
	libvirtd.

	Use /etc/init.d/libvirt-guests to manage clients on restart/shutdown of
	the host. The default configuration will suspend and resume running kvm
	guests with 'managedsave'. This behavior can be changed under
	/etc/conf.d/libvirt-guests


For systemd users:

	The '--listen' parameter is unavailable when libvirtd is run as a
    systemd unit.

	The configuration for the 'libvirt-guests.service' is found under
	/etc/libvirt/libvirt-guests.conf"


If you have built libvirt with policykit support, a new group "libvirt" has
been created. Simply add a user to the libvirt group in order to grant
administrative access to libvirtd. Alternatively, drop a custom policykit
rule into /etc/polkit-1/rules.d.

If you have built libvirt without policykit support (USE=-policykit), you
must change the unix sock group and/or perms in /etc/libvirt/libvirtd.conf
in order to allow normal users to connect to libvirtd.


If libvirt is built with USE=caps, libvirt will now start qemu/kvm VMs
with non-root privileges. Ensure any resources your VMs use are accessible
by qemu:qemu.