The bug with daemon-reload was fixed a long time ago.
The direct invocation of 'systemctl daemon-reload' with no redirect of
errors to /dev/null causes a user-visible message when installing into a chroot:
Running scriptlet: chrony-4.2-4.fc37.x86_64 320/334
System has not been booted with systemd as init system (PID 1). Can't operate.
Failed to connect to bus: Host is down
...
Add -F 2 to default /etc/sysconfig/chronyd to enable a filter blocking a
small number of specific system calls. The filter is incompatible with
the mailonchange directive.
The packaged NM dispatcher and dhclient scripts provide the same
functionality. Only one should be active depending on the system
configuration.
Instead of guessing in the NM dispatcher script whether the dhclient
script is executed, check in the dhclient script whether it is called
from a NM dispatcher script (the 11-dhclient script included in the
dhcp-client package).
This is simpler and it removes dependency on the 11-dhclient script.
Patch the upstream example `nm-dispatcher.dhcp` with
Fedora/RHEL-specific behavior including exiting if a dhclient
install is detected, and using the /etc/sysconfig paths.
Originally part of work for fixing integration between the
NetworkManager internal DHCP client and chrony:
https://bugzilla.redhat.com/show_bug.cgi?id=1800901
Drop the chrony-helper script in favor of the new sourcedir directive.
Modify the dhclient script to save NTP servers from DHCP to
/run/chrony-dhcp/$interface.sources and configure chronyd to (re)load
the sources from files in that directory on start and the "reload
sources" command.
Other functionality of the helper script is dropped with no
replacemement.
