add mon

4.monitoring/alerts.yaml

@@ -0,0 +1,41 @@
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  generation: 2
+  labels:
+    prometheus: zalando
+    role: alert-rules
+  name: prometheus-zalando-rules
+  namespace: zalando
+spec:
+  groups:
+    - name: zalando.rules
+      rules:
+        - alert: PG exporter
+          annotations:
+            description: >-
+              Failed to scrape {{ $labels.job }} on {{ $labels.namespace }} for
+              more than 2 minutes.
+            title: PG exporter is down
+          expr: up{job="zalando-monitoring"} == 0
+          for: 2m
+          labels:
+            severity: warning
+        - alert: PG InstanceDown
+          annotations:
+            description: Failed to scrape {{$labels.namespace}} for more than 3 minutes.
+            title: PG Instance is down
+          expr: pg_up == 0
+          for: 3m
+          labels:
+            severity: critical
+        - alert: PG Lag
+          annotations:
+            description: >-
+              Failed replication on replica {{$labels.namespace}} for more than
+              3 minutes.
+            title: PG Peplication lag
+          expr: pg_replication_slots_pg_wal_lsn_diff > 1024
+          for: 3m
+          labels:
+            severity: critical

4.monitoring/podMonitor.yaml

@@ -0,0 +1,19 @@
+apiVersion: monitoring.coreos.com/v1
+kind: PodMonitor
+metadata:
+  labels:
+    app: psql-pg-exporter
+  name: zalando-pg-exporter
+  namespace: rm-pgsql
+spec:
+  podMetricsEndpoints:
+    - path: /metrics
+      port: exporter
+      relabelings:
+        - action: replace
+          replacement: rke-first-cluster
+          targetLabel: cluster
+      scheme: http
+  selector:
+    matchLabels:
+      application: spilo