Merge pull request #11472 from vrutkovs/3.11-redeploy-monitoring

Add playbooks to regenerate TLS certificates for openshift-monitoring

playbooks/openshift-monitoring/private/redeploy-certificates.yml

@@ -0,0 +1,125 @@
+---
+- name: Update monitoring certificates
+ hosts: oo_first_master
+ vars:
+ roles:
+ - lib_openshift
+ - openshift_facts
+ tasks:
+ - name: Remove TLS secret
+ oc_obj:
+ name: "{{ item }}"
+ kind: secret
+ state: absent
+ namespace: openshift-monitoring
+ with_items:
+ - alertmanager-main-tls
+ - grafana-tls
+ - kube-state-metrics-tls
+ - node-exporter-tls
+ - prometheus-k8s-tls
+
+ - name: Remove Prometheus pods
+ oc_obj:
+ selector: "app=prometheus"
+ kind: pod
+ state: absent
+ namespace: openshift-monitoring
+
+ - name: Verify that the Prometheus is running
+ oc_obj:
+ namespace: openshift-monitoring
+ kind: statefulset
+ state: list
+ name: prometheus-k8s
+ register: prometheus_ss
+ until:
+ - prometheus_ss.results.results[0].status.readyReplicas is defined
+ - prometheus_ss.results.results[0].status.readyReplicas > 0
+ retries: 60
+ delay: 10
+ changed_when: false
+
+ - name: Remove Alertmanager pods
+ oc_obj:
+ selector: "app=alertmanager"
+ kind: pod
+ state: absent
+ namespace: openshift-monitoring
+
+ - name: Verify that the Alertmanager is running
+ oc_obj:
+ namespace: openshift-monitoring
+ kind: statefulset
+ state: list
+ name: alertmanager-main
+ register: alertmanager_ss
+ until:
+ - alertmanager_ss.results.results[0].status.readyReplicas is defined
+ - alertmanager_ss.results.results[0].status.readyReplicas > 0
+ retries: 60
+ delay: 10
+ changed_when: false
+
+ - name: Remove kube-state-metrics pods
+ oc_obj:
+ selector: "app=kube-state-metrics"
+ kind: pod
+ state: absent
+ namespace: openshift-monitoring
+
+ - name: Verify that the Kube State Metrics is running
+ oc_obj:
+ namespace: openshift-monitoring
+ kind: deployment
+ state: list
+ name: kube-state-metrics
+ register: kube_state_metrics_deployment
+ until:
+ - kube_state_metrics_deployment.results.results[0].status.readyReplicas is defined
+ - kube_state_metrics_deployment.results.results[0].status.readyReplicas > 0
+ retries: 60
+ delay: 10
+ changed_when: false
+
+ - name: Remove grafana pods
+ oc_obj:
+ selector: "app=grafana"
+ kind: pod
+ state: absent
+ namespace: openshift-monitoring
+
+ - name: Verify that the Grafana is running
+ oc_obj:
+ namespace: openshift-monitoring
+ kind: deployment
+ state: list
+ name: grafana
+ register: grafana_deployment
+ until:
+ - grafana_deployment.results.results[0].status.readyReplicas is defined
+ - grafana_deployment.results.results[0].status.readyReplicas > 0
+ retries: 60
+ delay: 10
+ changed_when: false
+
+ - name: Remove node-exporter pods
+ oc_obj:
+ selector: "app=node-exporter"
+ kind: pod
+ state: absent
+ namespace: openshift-monitoring
+
+ - name: Verify that the node-exporter is running
+ oc_obj:
+ namespace: openshift-monitoring
+ kind: daemonset
+ state: list
+ name: node-exporter
+ register: node_exporter_ds
+ until:
+ - node_exporter_ds.results.results[0].status.numberReady is defined
+ - node_exporter_ds.results.results[0].status.numberReady > 0
+ retries: 60
+ delay: 10
+ changed_when: false

playbooks/openshift-monitoring/redeploy-certificates.yml

@@ -0,0 +1,4 @@
+---
+- import_playbook: ../init/main.yml
+
+- import_playbook: private/redeploy-certificates.yml

playbooks/redeploy-certificates.yml

@@ -30,3 +30,6 @@
 
 - import_playbook: openshift-console/private/redeploy-certificates.yml
  when: openshift_console_install | default(true) | bool
+
+- import_playbook: openshift-monitoring/private/redeploy-certificates.yml
+ when: openshift_cluster_monitoring_operator_install | default(true) | bool