Use notify for workaround controllers unit.

roles/openshift_master/handlers/main.yml

@@ -7,8 +7,6 @@
   service: name={{ openshift.common.service_type }}-master-api state=restarted
   when: (openshift_master_ha | bool) and (not master_api_service_status_changed | default(false)) and openshift.master.cluster_method == 'native'
 
-# TODO: need to fix up ignore_errors here
 - name: restart master controllers
   service: name={{ openshift.common.service_type }}-master-controllers state=restarted
   when: (openshift_master_ha | bool) and (not master_controllers_service_status_changed | default(false)) and openshift.master.cluster_method == 'native'
-  ignore_errors: yes

roles/openshift_master/templates/atomic-openshift-master-controllers.service.j2

@@ -7,7 +7,7 @@ Before={{ openshift.common.service_type }}-node.service
 Requires=network.target
 
 [Service]
-Type=simple
+Type=notify
 EnvironmentFile=/etc/sysconfig/{{ openshift.common.service_type }}-master-controllers
 Environment=GOTRACEBACK=crash
 ExecStart=/usr/bin/openshift start master controllers --config=${CONFIG_FILE} $OPTIONS