add etcdbackup alerts

prometheus/alerts.yaml

@@ -89,6 +89,22 @@ spec:
         annotations:
           summary: "SMARTcheck not running"
           description: 'The last SMARTcheck on server {{ $labels.node }} was more than 3h ago. Plox fix.'
+    - name: etcdbackup
+      rules:
+      - alert: "etcdbackup too old"
+        expr: (time() - etcdbackup_time) > 10800
+        labels:
+          severity: warning
+        annotations:
+          summary: "etcd backup not running"
+          description: 'The last etcd backup on node {{ $labels.node }} was more than 3h ago. Plox fix.'
+      - alert: "etcdbackup failed"
+        expr: etcdbackup_result > 0
+        labels:
+          severity: warning
+        annotations:
+          summary: "etcdbackup failed"
+          description: "The backup script for etcd failed on node {{ $labels.node }}. Plox fix."
     - name: kubernetes
       rules:
       - alert: KubernetesUnhealthyPod