Restart octavia rsyslog after updating the config

After creating the octavia rsyslog configuration file, the
octavia_rsyslog container must be restarted if it's already running,
otherwise the new configuration is not applied in the container.

This issue is new, the octavia-ansible playbook was moved from step 5 to
post_deploy, which means that the file is now created after the first start
of the container.

Change-Id: Ibb439d47d551adc5af2d87234b732565af1cc90f
(cherry picked from commit 0a6f775201)
(cherry picked from commit 5d022f1c66)

tripleo_ansible/roles/octavia_controller_post_config/tasks/main.yml

@@ -41,6 +41,23 @@
         dest: "{{ octavia_confd_prefix }}/etc/rsyslog.d/10-octavia.conf"
         selevel: s0
         setype: svirt_sandbox_file_t
+
+    - name: check if octavia rsyslog is running
+      become: true
+      become_user: root
+      shell: |
+        systemctl is-active --quiet tripleo_octavia_rsyslog
+      failed_when: false
+      register: octavia_rsyslog_service
+
+    - name: restart octavia rsyslog service
+      become: true
+      become_user: root
+      systemd:
+        name: tripleo_octavia_rsyslog
+        state: restarted
+      when:
+        - octavia_rsyslog_service.rc == 0
   when: enable_log_offloading | bool
 
 - name: read the current IP list