fix: correct dashboard verification checks and retry helm upgrade lock

.gitea/workflows/dashboards.yml

@@ -93,6 +93,7 @@ jobs:
       - name: Verify Grafana content resources
         working-directory: ansible
         run: |
-          ansible -i inventory.ini 'control_plane[0]' -m command -a "kubectl -n observability get configmap | grep grafana"
+          ansible -i inventory.ini 'control_plane[0]' -m command -a "kubectl -n observability get configmap -l grafana_datasource=1"
+          ansible -i inventory.ini 'control_plane[0]' -m command -a "kubectl -n observability get configmap -l grafana_dashboard=1"
         env:
           ANSIBLE_HOST_KEY_CHECKING: "False"

ansible/roles/observability/tasks/main.yml

@@ -50,6 +50,10 @@
     --values /tmp/kube-prometheus-stack-values.yaml
     --wait
     --timeout 10m
+  register: kube_prom_install
+  retries: 6
+  delay: 20
+  until: kube_prom_install.rc == 0
   changed_when: true
 
 - name: Wait for Grafana deployment rollout