Collect status of all services

Would be helpful in troubleshooting services
which either fails to start or takes time to
start.

Related-Bug: #1970679
Change-Id: Iba2fce5f8b1cd00708f092e6eb5a1fbd96e97da0

Author: karelyatin

.zuul.yaml

@@ -398,6 +398,7 @@
         '{{ stage_dir }}/rpm-qa.txt': logs
         '{{ stage_dir }}/core': logs
         '{{ stage_dir }}/listen53.txt': logs
+        '{{ stage_dir }}/services.txt': logs
         '{{ stage_dir }}/deprecations.log': logs
         '{{ stage_dir }}/audit.log': logs
         /etc/ceph: logs

roles/capture-system-logs/README.rst

@@ -9,6 +9,7 @@ Stage a number of different logs / reports:
 - coredumps
 - dns resolver
 - listen53
+- services
 - unbound.log
 - deprecation messages
 

roles/capture-system-logs/tasks/main.yaml

@@ -19,6 +19,9 @@
           rpm -qa | sort > {{ stage_dir }}/rpm-qa.txt
       fi
 
+      # Services status
+      sudo systemctl status --all > services.txt 2>/dev/null
+
       # NOTE(kchamart) The 'audit.log' can be useful in cases when QEMU
       # failed to start due to denials from SELinux — useful for CentOS
       # and Fedora machines.  For Ubuntu (which runs AppArmor), DevStack