Skip to main content

Single-Step Sensor Installation Guide

This guide covers the steps to install and configure sensors for a Kubernetes cluster using a single-step installation process.

Step 1: Create Configuration File​

Create a file named config.yaml with the following structure. Update values as needed for your environment.

# config.yaml
KUBECOL_HOST: 127.0.0.1
AGGREGATOR_HOST: kubesense.ai
AGGREGATOR_PORT: "30054"
METRICS_STORE_HOST: kubesense.ai
METRICS_STORE_PORT: "30060"
SENSOR_INSTALLATION_TYPE: legacy
DOCKER_COMPOSE_PROJECT_NAME: kubesense
SENSOR_GROUP_ID: xxx
CLUSTER_NAME: xxx
KUBESENSOR_AUTH: xxx
LOG_PATHS:
  # Format: service_name: log_file_path

Step 2: Run Installation Script​

Pass the configuration file to the install script using the following command:

bash <(wget -qO- https://docker.kubesense.ai/setup.sh) sensor config.yaml

This script deploys and configures all required services and components based on the values defined in config.yaml.

Step 3: APM Instrumentation (Node.js)​

Below is the example for Node.js application instrumentation.

Install OpenTelemetry Packages​

npm install --save @opentelemetry/api
npm install --save @opentelemetry/auto-instrumentations-node

Set Environment Variables​

Configure your application's environment with the following variables:

env: {
    OTEL_SERVICE_NAME: "", // Set your service name here
    OTEL_RESOURCE_ATTRIBUTES: "kubesense.env_type=legacy,kubesense.cluster=xxx",
    OTEL_EXPORTER_OTLP_TRACES_ENDPOINT: "http://kubesense.ai:33443/v1/traces",
    OTEL_NODE_RESOURCE_DETECTORS: "env,host,os,process",
    NODE_OPTIONS: "--require @opentelemetry/auto-instrumentations-node/register",
}

These settings enable automatic instrumentation and export traces to the configured Kubesense OTLP endpoint.

Additional Information​

  • Required Permissions: The container runs in privileged mode to enable kernel-level operations like eBPF.

For more details, visit KubeSense Documentation.