Description of problem:

Tracer will normally display the commands to restart services. Unfortunately, the logic seems limited and some running services are just mentioned in the section of programs to restart manually, leaving it up to the system manager to figure out how to restart them.

Version-Release number of selected component (if applicable):
tracer-common-0.6.12-4.fc25.noarch


How reproducible:


Steps to Reproduce:
1. Install updates
2. Run 'tracer -r' (or use the dnf plugin)

Actual results:

# tracer -r
You should restart:
  * Some applications using:
      systemctl restart NetworkManager
      systemctl restart accounts-daemon
      systemctl restart atd
      systemctl restart auditd
      systemctl restart crond
      systemctl restart httpd
      systemctl restart irqbalance
      systemctl restart lightdm
      systemctl restart smartd
      systemctl restart sshd
      systemctl restart sssd

  * These applications manually:
      Xorg
      abrt-dbus
      automount
      cupsd
      rsyslogd
      sedispatch
      sssd_be
      sssd_nss
      sssd_pam
      x2gocleansessio


Expected results:
to see all restartable services listed with their systemctl command.

Additional info:

It seems tracer cannot handle the cases where the running daemon doesn't have the exact same name as its systemd service or unit (automount vs autofs, cupsd vs cups), or when a process is actually a subprocess of another daemon (Xorg started from lightdm, and the sssd subprocesses).

Perhaps it would be possible to add some logic here, such as a translation table, or to ask systemd for the master process (perhaps using cgroup?).