okd/roles/openshift_logging_elasticsearch/tasks/restart_cluster.yml

---
## get all pods for the cluster
- command: >
    {{ openshift_client_binary }}
    --config={{ openshift.common.config_base }}/master/admin.kubeconfig
    get pod
    -l component={{ _cluster_component }},provider=openshift
    -n {{ openshift_logging_elasticsearch_namespace }}
    -o jsonpath={.items[?(@.status.phase==\"Running\")].metadata.name}
  register: _cluster_pods

# make a temp dir for admin certs
- command: mktemp -d /tmp/openshift-logging-ansible-XXXXXX
  register: _logging_handler_tempdir
  changed_when: False
  check_mode: no

- name: Exporting secrets to use communicating with the ES cluster
  command: >
    {{ openshift_client_binary }}
    --config={{ openshift.common.config_base }}/master/admin.kubeconfig
    extract secret/logging-elasticsearch
    -n {{ openshift_logging_elasticsearch_namespace }}
    --keys=admin-cert --keys=admin-key
    --to={{ _logging_handler_tempdir.stdout }}

### Check for cluster state before making changes -- if its red, yellow or missing nodes then we don't want to continue
- name: "Checking current health for {{ _es_node }} cluster"
  command: >
    curl -s -k
    --cert {{ _logging_handler_tempdir.stdout }}/admin-cert
    --key {{ _logging_handler_tempdir.stdout }}/admin-key
    https://logging-{{ _cluster_component }}.{{ openshift_logging_elasticsearch_namespace }}.svc:9200/_cluster/health?pretty
  register: _pod_status
  when:
    - _cluster_pods.stdout
    - _cluster_pods.stdout.split(' ') | count > 0

- when:
    - _pod_status.stdout is defined
    - (_pod_status.stdout | from_json)['status'] in ['yellow', 'red'] or (_pod_status.stdout | from_json)['number_of_nodes'] != _cluster_pods.stdout.split(' ') | count
  block:
    - name: Set Logging message to manually restart
      run_once: true
      set_stats:
        data:
          installer_phase_logging:
            message: "Cluster logging-{{ _cluster_component }} was not in an optimal state and will not be automatically restarted. Please see documentation regarding doing a {{ 'full' if full_restart_cluster | bool else 'rolling'}} cluster restart."

    - debug: msg="Cluster logging-{{ _cluster_component }} was not in an optimal state and will not be automatically restarted. Please see documentation regarding doing a {{ 'full' if full_restart_cluster | bool else 'rolling'}} cluster restart."

- when: _pod_status.stdout is undefined or ( (_pod_status.stdout | from_json)['status'] in ['green'] and (_pod_status.stdout | from_json)['number_of_nodes'] == _cluster_pods.stdout.split(' ') | count )
  block:
    - command: >
        {{ openshift_client_binary }}
        --config={{ openshift.common.config_base }}/master/admin.kubeconfig
        get dc
        -l component={{ _cluster_component }},provider=openshift
        -n {{ openshift_logging_elasticsearch_namespace }}
        -o jsonpath={.items[*].metadata.name}
      register: _cluster_dcs

    ## restart all dcs for full restart
    - name: "Performing full cluster restart for {{ _cluster_component }} cluster"
      include_tasks: full_cluster_restart.yml
      vars:
        logging_restart_cluster_dcs: "{{ _cluster_dcs.stdout_lines }}"
      when:
        - full_restart_cluster | bool

    ## restart the node if it's dc is in the list of nodes to restart
    - name: "Performing rolling cluster restart for {{ _cluster_component }} cluster"
      include_tasks: rolling_cluster_restart.yml
      vars:
        logging_restart_cluster_dcs: "{{ _restart_logging_nodes | intersect(_cluster_dcs.stdout) }}"
      when:
        - not full_restart_cluster | bool

# remove temp dir
- name: Cleaning up temp dir
  file:
    path: "{{ _logging_handler_tempdir.stdout }}"
    state: absent
  changed_when: False