New validation: detect failed containers

Failed containers are pretty bad, since we have a degraded service.

Running this validation before an upgrade is a good thing, and
running it after a deploy/upgrade will ensure we're in a right state

Co-Authored-by: Gaël Chamoulaud <gchamoul@redhat.com>
Change-Id: I242f1c7cff76e8304696ea10b32c1545fa5b8ea5

doc/source/roles/role-container-status.rst

@@ -0,0 +1,6 @@
+================
+container-status
+================
+
+.. ansibleautoplugin::
+   :role: roles/container-status

playbooks/container-status.yaml

@@ -0,0 +1,13 @@
+---
+- hosts: undercloud, overcloud
+  vars:
+    metadata:
+      name: Ensure container status
+      description: >
+        Detect failed containers and raise an error.
+      groups:
+        - pre-upgrade
+        - post-deployment
+        - post-upgrade
+  roles:
+    - container-status

roles/container-status/tasks/main.yaml

@@ -0,0 +1,45 @@
+---
+- name: Set oc_container_cli fact for the Overcloud nodes
+  set_fact:
+    oc_container_cli: "{{ hostvars[inventory_hostname].container_cli | default('podman', true) }}"
+  when:
+    - "'overcloud' in group_names"
+    - oc_container_cli is not defined
+
+- when: "'Undercloud' in group_names"
+  block:
+  - name: Set container_cli fact from undercloud.conf
+    block:
+    - name: Get the path of tripleo undercloud config file
+      become: true
+      hiera:
+        name: "tripleo_undercloud_conf_file"
+
+    - name: Get container client from undercloud.conf
+      ini:
+        path: "{{ tripleo_undercloud_conf_file }}"
+        section: DEFAULT
+        key: container_cli
+        ignore_missing_file: true
+      register: container_cli
+
+    - name: Set uc_container_cli for the Undercloud
+      set_fact:
+        uc_container_cli: "{{ container_cli.value|default('podman', true) }}"
+      when: uc_container_cli is not defined
+
+- name: Get failed containers for podman
+  changed_when: false
+  become: True
+  command: >
+    {% if oc_container_cli is defined %}{{ oc_container_cli }}{% else %}{{ uc_container_cli }}{% endif %}
+    {% raw %}
+    ps -a --filter 'status=exited' --format '{{ .Names }} {{ .Status }}'
+    {% endraw %}
+  register: failed_containers
+
+- name: Fail if we detect failed containers
+  fail:
+    msg: "Failed container detected: {{ item }}."
+  when: item is not match(".* Exited \(0\) .* ago")
+  loop: "{{ failed_containers.stdout_lines }}"