Skip to main content
Version: 3.0.0-alpha (Diátaxis)

Monitoring Agents

monitoringAgents (Object) — Required​

Description​

The monitoringAgents field contains the main configuration for the monitoring agents deployed across the cluster. It allows enabling the component and customizing its behavior through Helm values.

Example​

monitoringAgents:
  enabled: true
  valuesOverride:
    monitoringAgents:
      nodeExporter:
        enabled: true

enabled (boolean) — Required​

Description​

Indicates whether monitoring agents are enabled (true) or disabled (false). When enabled, the agents automatically collect system and Kubernetes metrics and export them to a monitoring backend (e.g., Prometheus, Grafana Agent, OpenTelemetry Collector, etc.).

Example​

enabled: true

valuesOverride (Object) — Required​

Description​

The valuesOverride field allows overriding the Helm values used to deploy monitoring agents. It is used to adjust configuration (module activation, resources, listening ports, labels, etc.).

Example​

Monitoring Agents​

Log and metrics collection using FluentBit and other agents.

valuesOverride:
  # FluentBit configuration
  fluentbit:
    enabled: true
    config:
      outputs: |
        [OUTPUT]
            Name forward
            Match *
            Host fluent-aggregator.logging
            Port 24224

Advanced Monitoring Configuration​

valuesOverride:
  monitoringAgents:
    # FluentBit for logs
    fluentbit:
      enabled: true
      resources:
        requests:
          cpu: 5m
          memory: 10Mi
        limits:
          cpu: 50m
          memory: 60Mi
      config:
        service: |
          [SERVICE]
              Flush         1
              Log_Level     info
              Daemon        off
              Parsers_File  parsers.conf
        inputs: |
          [INPUT]
              Name              tail
              Path              /var/log/containers/*.log
              Parser            cri
              Tag               kube.*
              Refresh_Interval  5
              Mem_Buf_Limit     50MB
        outputs: |
          [OUTPUT]
              Name  forward
              Match *
              Host  logs.company.com
              Port  24224

    # Node Exporter for system metrics
    nodeExporter:
      enabled: true
      resources:
        requests:
          cpu: 10m
          memory: 32Mi
        limits:
          cpu: 200m
          memory: 128Mi

💡 Best Practices​

  • Enable enabled: true to ensure continuous collection of system and application metrics.
  • Use valuesOverride to fine-tune agent configuration depending on the environment (e.g., limit collection on specific nodes).
  • Configure appropriate resource limits to prevent agents from impacting cluster load.