main.yml

---
prometheus_version: 2.0.0

prometheus_config_dir: /etc/prometheus
prometheus_db_dir: /var/lib/prometheus
prometheus_root_dir: /opt/prometheus

prometheus_web_listen_address: "0.0.0.0:9090"
prometheus_web_external_url: ''

prometheus_storage_retention: "30d"

prometheus_config_flags_extra: {}
#prometheus_config_flags_extra:
#  storage.tsdb.retention: 15d
#  alertmanager.timeout: 10s

prometheus_alertmanager_config: []
#prometheus_alertmanager_config:
#  - scheme: https
#    path_prefix: /alertmanager
#    basic_auth:
#      username: user
#      password: pass
#    static_configs:
#      - targets: ["127.0.0.1:9093"]
#    proxy_url: "127.0.0.2"

prometheus_global:
  scrape_interval: 60s
  scrape_timeout: 15s
  evaluation_interval: 15s

prometheus_external_labels:
  environment: "{{ ansible_fqdn | default(ansible_host) | default(inventory_hostname) }}"

prometheus_targets:
  node:
  - targets:
    - localhost:9100
    labels:
      env: test

prometheus_scrape_configs:
- job_name: "prometheus"
  metrics_path: "{{ prometheus_metrics_path }}"
  static_configs:
  - targets:
    - "{{ ansible_fqdn | default(ansible_host) | default('localhost') }}:9090"
- job_name: "node"
  file_sd_configs:
  - files:
    - "{{ prometheus_config_dir }}/file_sd/node.yml"

# Alternative config file name, searched in ansible templates path.
prometheus_config_file: 'prometheus.yml.j2'

prometheus_alert_rules:
- alert: InstanceDown
  expr: "up == 0"
  for: 5m
  labels:
    severity: critical
  annotations:
    description: "{% raw %}{{ $labels.instance }} of job {{ $labels.job }} has been down for more than 5 minutes.{% endraw %}"
    summary: "{% raw %}Instance {{ $labels.instance }} down{% endraw %}"
- alert: CriticalCPULoad
  expr: '(100 * (1 - avg(irate(node_cpu{job="node",mode="idle"}[5m])) BY (instance))) > 96'
  for: 2m
  labels:
    severity: critical
  annotations:
    description: "{% raw %}{{ $labels.instance }} of job {{ $labels.job }} has Critical CPU load for more than 2 minutes.{% endraw %}"
    summary: "{% raw %}Instance {{ $labels.instance }} - Critical CPU load{% endraw %}"
- alert: CriticalDiskSpace
  expr: 'node_filesystem_free{job="node",filesystem!~"^/run(/|$)"} / node_filesystem_size{job="node"} < 0.1'
  for: 4m
  labels:
    severity: critical
  annotations:
    description: "{% raw %}{{ $labels.instance }} of job {{ $labels.job }} has less than 10% space remaining.{% endraw %}"
    summary: "{% raw %}Instance {{ $labels.instance }} - Critical disk space usage{% endraw %}"
- alert: RebootRequired
  expr: "node_reboot_required > 0"
  labels:
    severity: warning
  annotations:
    descritpion: "{% raw %}{{ $labels.instance }} requires a reboot.{% endraw %}"
    summary: "{% raw %}Instance {{ $labels.instance }} - reboot required{% endraw %}"