grafana: make alerts better for fux

resources/chaosknoten/grafana/docker_compose/alertmanager.yaml.j2

@@ -79,7 +79,7 @@ receivers:
   - name: "email-fux-critical"
     email_configs:
       - send_resolved: true
-        to: "stb@lassitu.de"
+        to: "stb@lassitu.de,fux@zimdahl.org"
         from: "alert-manager@hamburg.ccc.de"
         smarthost: "cow.hamburg.ccc.de:587"
         auth_username: "alert-manager@hamburg.ccc.de"

resources/chaosknoten/grafana/docker_compose/compose.yaml.j2

@@ -14,6 +14,7 @@ services:
     volumes:
       - ./configs/prometheus.yml:/etc/prometheus/prometheus.yml
       - ./configs/prometheus_alerts.rules.yaml:/etc/prometheus/rules/alerts.rules.yaml
+      - ./configs/prometheus_alerts-fux.rules.yaml:/etc/prometheus/rules/alerts-fux.rules.yaml
       - prom_data:/prometheus
 
   alertmanager:

resources/chaosknoten/grafana/docker_compose/grafana-datasource.yml

@@ -18,4 +18,3 @@ datasources:
       httpHeaderName1: "X-Scope-OrgID"
     secureJsonData:
       httpHeaderValue1: "chaos"
-

resources/chaosknoten/grafana/docker_compose/prometheus_alerts-fux.rules.yaml

@@ -0,0 +1,41 @@
+groups:
+  - name: Fux-Generic
+    rules:
+      - alert: HostJobFlaky
+        expr: group by(instance, job) (changes(up{org="fux"}[24h]) > 7)
+        for: 0m
+        labels:
+          severity: info
+          org: fux
+        annotations:
+          summary: Job {{ $labels.job }} flaky on (instance {{ $labels.instance }})
+          description: "The job {{ $labels.job }} on target: {{ $labels.instance }} has been flaky over the last 24 hours."
+  - name: Fux-SNMP
+    rules:
+      - alert: SnmpTargetMissing
+        expr: up{job=~".*snmp.*", org="fux"} == 0
+        for: 15m
+        labels:
+          severity: critical
+          org: fux
+        annotations:
+          summary: SNMP target missing (instance {{ $labels.instance }})
+          description: "SNMP target: {{ $labels.instance }} has disappeared for more the 15 min."
+  - name: Fux-DHCP
+    rules:
+      - alert: DhcpFuxSharedFailed
+        expr: script_success{script="check_dhcp_fux_shared"} == 0
+        for: 0m
+        labels:
+          severity: critical
+        annotations:
+          summary: DHCP for Fux Shared stoped working
+          description: "No DHCP lease for the Fux Shared range was received \n V"
+      - alert: DhcpFuxAdminFailed
+        expr: script_success{script_success="check_dhcp_fux_admin"} == 0
+        for: 0m
+        labels:
+          severity: critical
+        annotations:
+          summary: DHCP for Fux Admin stoped working
+          description: "No DHCP lease for the Fux Admin range was received"

resources/chaosknoten/grafana/docker_compose/prometheus_alerts.rules.yaml

@@ -410,7 +410,7 @@ groups:
           summary: Prometheus job missing (instance {{ $labels.instance }})
           description: "A Prometheus job has disappeared\n  VALUE = {{ $value }}"
       - alert: PrometheusTargetMissing
-        expr: up == 0
+        expr: up{job!~"snmp|noc_room_temp"} == 0
         for: 0m
         labels:
           severity: critical
@@ -418,7 +418,7 @@ groups:
           summary: Prometheus target missing (instance {{ $labels.instance }})
           description: "A Prometheus target has disappeared. An exporter might be crashed.\n  VALUE = {{ $value }}"
       - alert: PrometheusAllTargetsMissing
-        expr: sum by (job) (up) == 0
+        expr: sum by (job) (up{job!~"snmp|noc_room_temp"}) == 0
         for: 0m
         labels:
           severity: critical