alert.rules.yaml

groups:
- name: "Load System"
  rules:

#       description: "{{ $labels.instance }} of job {{ $labels.job }} is under high load."

       # Alert for any instance that has a median request late
   - alert: high_treta_load
     expr: netdata_system_load_load_average{dimension!="load5",dimension!="load15"} > 2.0
     for: 10m
     labels:
       severity: atention
     annotations:
       summary: "High load system CPU"
       description: "{{ $labels.instance }} is under high load ({{ $value }})"


- name: "Disponibility"
  rules:

   - alert: service_down
     expr: up == 0
     for: 30s
     labels:
       severity: page
     annotations:
       summary: "Instance {{ $labels.instance }} down"
       description: "{{ $labels.instance }} lost connection whit API for more than {{ $value }}s."

- name: "File System Usage"
  rules:

   - alert: usage_treta_fs
     expr: netdata_disk_space_GB_average{dimension="avail",family="/"} < 5
     for: 3m
     annotations:
       summary: "Disk space usage for  /"
       description: "{{ $labels.instance }} current space avail is: ({{ $value }}g)"

   #- alert: LowRootFilesytem
   # expr: if ((node_filesystem_size{mountpoint="/root-disk"} - node_filesystem_free{mountpoint="/root-disk"} ) / node_filesystem_size{mountpoint="/root-disk"} * 100) > 75
   #  expr: if ((netdata_disk_space_GB_average{dimension="used",family="/"} * 100) / (netdata_disk_space_GB_average{dimension="avail",family="/"} + netdata_disk_space_GB_average{dimension="used",family="/"})
   #   for: 2m
   #   labels:
   #     severity:
   #   annotations:
   #     summary: "{{$labels.instance}}: Low root disk space",
   #     description: "{{$labels.instance}}: Root disk usage is above 75% (current value is: {{ $value  }})"

   - alert: Disk_Will_Fill_In_3Hours
     expr: predict_linear (netdata_disk_space_GB_average{dimension="avail",family="/"}[1h], 3 * 3600) < 0
     for: 5m
     labels:
       severity: page
     annotations:
       summary: "Filesystem / will be exhausted in 3 hours"
       description: "We recovered 1 hour of history to predict 3 hours ahead"