monitoring/variables.yml

---

instance: monitoring

vault:
  pki:
    path: '[[ .prometheus.vault_pki ]]'
    ou: Monitoring
  rand_secrets:
    - path: grafana
      fields:
        - secret_key
        - initial_admin_pwd

monitoring:

  namespaces:
    - '*'

  exporters:
    count: 1

    ping:
      version: 1.1.0
      image: '[[ .docker.repo ]]ping-exporter:[[ .monitoring.exporters.ping.version ]]-1'
      env: {}
      resources:
        cpu: 10
        memory: 25
      probes: []

    blackbox:
      version: 0.24.0
      image: '[[ .docker.repo ]]blackbox-exporter:[[ .monitoring.exporters.blackbox.version ]]-1'
      env: {}
      resources:
        cpu: 10
        memory: 50
      tcp_probes: []
      http_probes: []

    consul:
      version: 0.11.0
      image: '[[ .docker.repo ]]consul-exporter:[[ .monitoring.exporters.consul.version ]]-2'
      env: {}
      resources:
        cpu: 20
        memory: 32
      vault:
        policies:
          - 'consul-exporter[[ .consul.suffix ]]'

    cluster:
      image: nginxinc/nginx-unprivileged:alpine
      env: {}
      resources:
        cpu: 10
        memory: 15
      vault:
        policies:
          - 'cluster-exporter[[ .consul.suffix ]]'
          - metrics

  prometheus:

    version: 2.51.0

    count: 1

    image: '[[ .docker.repo ]]prometheus:[[ .monitoring.prometheus.version ]]-1'

    env: {}

    resources:
      cpu: 200
      memory: 512

    volumes:
      data:
        type: csi
        source: 'prometheus-data'
        per_alloc: true

    vault:
      policies:
        - 'prometheus[[ .consul.suffix ]]'

    jobs: {}
    alert_rules: {}
    # alert_rules:
    #   postgres:
    #     url: https://raw.githubusercontent.com/samber/awesome-prometheus-alerts/master/dist/rules/prometheus-self-monitoring/embedded-exporter.yml

    public_url: https://prometheus.example.org
    traefik:
      enabled: true
      router: prometheus

    retention: 30d

    prometheus:
      enabled: true
      metrics_url: http://localhost:9090/metrics

  alertmanager:
    count: 1
    version: 0.27.0
    image: '[[ .docker.repo ]]alertmanager:[[ .monitoring.alertmanager.version ]]-1'
    env: {}
    resources:
      cpu: 50
      memory: 64
      memory_max: 80
    public_url: https://alerte.example.org
    traefik:
      enabled: true
      router: alertmanager
      strip_prefix: false
    volumes:
      data:
        source: 'alertmanager-data'
        type: csi
        per_alloc: true
    prometheus:
      metrics_url: http://127.0.0.1:9093/metrics
    vault:
      policies:
        - metrics
        - 'alertmanager[[ .consul.suffix ]]'
    email:
      from: alertmanager@[[ .consul.domain ]]
    custom_config: {}

  loki:
    version: 2.9.6
    image: '[[ .docker.repo ]]loki:[[ .monitoring.loki.version ]]-1'
    env: {}
    resources:
      cpu: 150
      memory: 512
    vault:
      policies:
        - 'loki[[ .consul.suffix ]]'
    public_url: https://loki.example.org
    traefik:
      router: loki
    retention: 720h # 1 month
    custom_config: {}
    prometheus:
      metrics_url: http://localhost:3100/metrics
    volumes:
      data:
        type: csi
        source: 'loki-data'

  vector:
    version: 0.36.1
    image: '[[ .docker.repo ]]vector:[[ .monitoring.vector.version ]]-1'

  aggregator:
    count: 1
    image: '[[ .monitoring.vector.image ]]'
    env: {}
    resources:
      cpu: 100
      memory: 192
    consul:
      connect:
        upstreams:
          - destination_name: 'loki[[ .consul.suffix ]]'
            local_bind_port: 3100
    vault:
      policies:
        - metrics[[ .consul.suffix ]]
    fluentd:
      enabled: false
      traefik:
        router: fluentd
        entrypoints:
          - fluentd
    syslog_udp:
      enabled: false
      traefik:
        router: syslog-udp
        entrypoints:
          - syslog
    vector:
      enabled: true
      public_url: https://vector.example.org
      traefik:
        enabled: false
    prometheus:
      metrics_url: http://127.0.0.1:9001/metrics

  grafana:
    version: 10.4.1
    image: '[[ .docker.repo ]]grafana:[[ .monitoring.grafana.version ]]-1'
    env:
      GF_SECURITY_ADMIN_PASSWORD: '{{ with secret "[[ .vault.root ]]kv/service/[[ .instance ]]/grafana" }}{{ .Data.data.initial_admin_pwd }}{{ end }}'
    resources:
      cpu: 100
      memory: 256
    public_url: https://grafana.example.org
    plugins:
      #- alexanderzobnin-zabbix-app
      #- ddurieux-glpi-app
      - grafana-clock-panel
      - grafana-piechart-panel
    feature_toggles: {}
    traefik:
      enabled: true
      router: grafana
      strip_prefix: false
    consul:
      connect:
        upstreams:
          - destination_name: postgres[[ .consul.suffix ]]
            local_bind_port: 5432
          - destination_name: loki[[ .consul.suffix ]]
            local_bind_port: 3100
          - destination_name: prometheus[[ .consul.suffix ]]
            local_bind_port: 9090
    volumes:
      data:
        type: csi
        source: 'grafana-data'
    vault:
      policies:
        - 'grafana[[ .consul.suffix ]]'
      database:
        role: grafana
        pgrole: grafana
    postgres:
      database: grafana
      user: '{{ with secret "[[ .vault.root ]]database/creds/grafana" }}{{ .Data.username }}{{ end }}'
      password: '{{ with secret "[[ .vault.root ]]database/creds/grafana" }}{{ .Data.password }}{{ end }}'
      pooler:
        mode: session
    prometheus:
      metrics_url: http://localhost:3000[[ (urlParse .monitoring.grafana.public_url).Path ]]/metrics

  agent:
    consul:
      meta:
        alloc: '${node.unique.name}'
    nomad:
      node_pool: all
    nomad_vector_logger:
      version: 24.3
      image: '[[ .docker.repo ]]nomad-vector-logger:[[ .monitoring.agent.nomad_vector_logger.version ]]-2'
      env: {}
      resources:
        cpu: 20
        memory: 24
        memory_max: 50
      vault:
        policies:
          - nomad-vector-logger[[ .consul.suffix ]]
    vector:
      image: '[[ .monitoring.vector.image ]]'
      env: {}
      resources:
        cpu: 100
        memory: 192
        memory_max: 384
      vault:
        policies:
          - metrics[[ .consul.suffix ]]
      consul:
        connect:
          upstreams:
            - destination_name: loki[[ .consul.suffix ]]
              local_bind_port: 3100
      volumes:
        nomad:
          type: host
          source: nomad_alloc
          read_only: true
        data:
          type: host
          source: vector_data
      prometheus:
        metrics_url: http://127.0.0.1:9001/metrics

prometheus:
  enabled: true