🚀 loki

kubernetes/apps/monitoring/loki/app/rules/loki-alerting-rules.yaml

@@ -0,0 +1,79 @@
+---
+groups:
+  - name: smart
+    rules:
+      - alert: SMARTFailure
+        expr: |
+          sum by (hostname) (count_over_time({hostname=~".+"} | json | _SYSTEMD_UNIT = "smartmontools.service" !~ "(?i)previous self-test completed without error" !~ "(?i)Prefailure" |~ "(?i)(error|fail)"[2m])) > 0
+        for: 2m
+        labels:
+          severity: critical
+          category: logs
+        annotations:
+          hostname: "{{ $labels.hostname }}"
+          summary: "{{ $labels.hostname }} has reported SMART failures"
+
+  - name: zigbee2mqtt
+    rules:
+      - alert: ZigbeeMQTTUnreachable
+        expr: |
+          sum(count_over_time({app="zigbee2mqtt"} |~ "(?i)not connected to mqtt server"[2m])) > 0
+        for: 2m
+        labels:
+          severity: critical
+          category: logs
+        annotations:
+          app: "{{ $labels.app }}"
+          summary: "{{ $labels.app }} is unable to reach MQTT"
+
+  - name: zwave-js-ui
+    rules:
+      - alert: ZwaveMQTTUnreachable
+        expr: |
+          sum(count_over_time({app="zwave-js-ui"} |~ "(?i)error while connecting mqtt"[2m])) > 0
+        for: 2m
+        labels:
+          severity: critical
+          category: logs
+        annotations:
+          app: "{{ $labels.app }}"
+          summary: "{{ $labels.app }} is unable to reach MQTT"
+
+  - name: frigate
+    rules:
+      - alert: FrigateMQTTUnreachable
+        expr: |
+          sum(count_over_time({app="frigate"} |~ "(?i)unable to connect to mqtt server"[2m])) > 0
+        for: 2m
+        labels:
+          severity: critical
+          category: logs
+        annotations:
+          app: "{{ $labels.app }}"
+          summary: "{{ $labels.app }} is unable to reach MQTT"
+
+  - name: home-assistant
+    rules:
+      - alert: HomeAssistantPostgresUnreachable
+        expr: |
+          sum by (app) (count_over_time({app="home-assistant"} |~ "(?i)error in database connectivity"[2m])) > 0
+        for: 2m
+        labels:
+          severity: critical
+          category: logs
+        annotations:
+          app: "{{ $labels.app }}"
+          summary: "{{ $labels.app }} is unable to connect to postgres"
+
+  - name: bazarr
+    rules:
+      - alert: BazarrJobRaisedException
+        expr: |
+          sum by (app) (count_over_time({app="bazarr"} |~ "(?i)Job(.+)Update(.+)from(.+)raised an exception"[2m])) > 0
+        for: 2m
+        labels:
+          severity: warning
+          category: logs
+        annotations:
+          app: "{{ $labels.app }}"
+          summary: "{{ $labels.app }} is raising job exceptions"