alertingrules_example.yaml

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  labels:
    prometheus: basic_rules_prom
    role: alert-rules
  name: application-basic-pack
  namespace: prometheus-platform-instances
spec:
  groups:
  - name: ./applications.basic.rules
    interval: 30s
    rules:
    - alert: "Application waiting"
      labels:
        application: "{{ $labels.exported_container }}"
      for: "5m"
      expr: "sum without(instance, exported_pod, endpoint, job)(kube_pod_container_status_waiting)>0"
      annotations:
        priority: P5
        description: "{{ $labels.exported_container }} is not able to start in the {{ $labels.exported_namespace }} system"
    - alert: "Application network receive errors"
      labels:
        application: "{{ $labels.pod }}"
      for: "5m"
      expr: "sum by(pod)(rate(container_network_receive_errors_total[10m])>0)"
      annotations:
        priority: P2
        description: "{{ $labels.pod}}"
    - alert: "Application network transmit errors"
      labels:
        application: "{{ $labels.pod }}"
      for: "5m"
      expr: "(sum by (pod) (rate(container_network_transmit_errors_total[10m])))>0"
      annotations:
        priority: P2
        description: "{{ $labels.pod}}"
    - alert: "Application responding to slow to http requests"
      labels:
        application: "{{ $labels.ingress }}"
      for: "5m"
      expr: "(max(rate(nginx_ingress_controller_request_duration_seconds_sum[10m])) by (exported_namespace, ingress))>2"
      annotations:
        priority: P2
        description: "Calls to {{ $labels.ingress }} in the {{ $labels.exported_namespace }} system is taking more than 2 second to respond"
    - alert: "Application API endpoint is failing"
      labels:
        application: "{{ $labels.exported_service }}"
      for: "5m"
      expr: "(sum(increase(nginx_ingress_controller_requests{status=~'^5.*'}[10m])) by (exported_service, status))>0"
      annotations:
        priority: P2
        description: "Calls to {{ $labels.exported_service }} is returning {{ $labels.status }}"
    - alert: "Application low on Memory"
      labels:
        application: "{{ $labels.exported_container }}"
      for: "10m"
      expr: "((sum by (exported_container) (label_replace(rate(container_memory_usage_bytes[5m]),'exported_container', '$1', 'container', '(.*)')))/(sum by (exported_container) (kube_pod_container_resource_limits{resource='memory'})))>0.8"
      annotations:
        priority: P2
        description: "{{ $labels.exported_container }} is running low on Memory"
    - alert: "Application low on CPU"
      labels:
        application: "{{ $labels.exported_container }}"
      for: "10m"
      expr: "((sum by (exported_container) (label_replace(rate(container_cpu_usage_seconds_total[5m]),'exported_container', '$1', 'container', '(.*)')))/(sum by (exported_container) (kube_pod_container_resource_limits{resource='cpu'})))>0.8"
      annotations:
        priority: P2
        description: "{{ $labels.exported_container }} is running low on CPU"