serverFiles:
  alerts:
    groups:
    - name: k8s_alert_rules
      rules:
      # ALERT when container memory usage exceed 90%
      - alert: container_mem_over_90
        expr: (sum(container_memory_working_set_bytes{image!="",name=~"^k8s_.*", pod_name!=""}) by (pod_name)) / (sum (container_spec_memory_limit_bytes{image!="",name=~"^k8s_.*", pod_name!=""}) by (pod_name)) > 0.9 and (sum(container_memory_working_set_bytes{image!="",name=~"^k8s_.*", pod_name!=""}) by (pod_name)) / (sum (container_spec_memory_limit_bytes{image!="",name=~"^k8s_.*", pod_name!=""}) by (pod_name)) < 2
        for: 2m
        annotations:
          summary: "{{ $labels.pod_name }}'s memory usage alert"
          description: "Memory Usage of Pod {{ $labels.pod_name }} on {{ $labels.kubernetes_io_hostname }} has exceeded 90% for more than 2 minutes."

      # ALERT when node is down
      - alert: node_down
        expr: up == 0
        for: 60s
        annotations:
          summary: "Node {{ $labels.kubernetes_io_hostname }} is down"
          description: "Node {{ $labels.kubernetes_io_hostname }} is down"