Setup

Installation and Setup¶

The OpenTelemetry Collector is a critical component in observability pipelines, aggregating and processing telemetry data from applications and services. This section guides you through installing and configuring the Collector on various platforms, ensuring it is ready to collect, process, and export traces, metrics, and logs.

Installation Methods¶

1. Docker¶

The Docker method is ideal for quick deployment and testing. Pull the official image and run the Collector with a configuration file:

docker pull otelcol/otelcol
docker run --rm -v $(pwd)/config.yaml:/etc/otelcol/config.yaml -p 13133:13133 otelcol/otelcol --config /etc/otelcol/config.yaml

Replace config.yaml with your actual configuration file. The Collector will expose metrics on port 13133 by default.

2. Package Managers¶

Install via your OS's package manager. For example, on Ubuntu/Debian:

sudo apt-get install -y apt-transport-https
curl -sSL https://raw.githubusercontent.com/open-telemetry/opentelemetry-collector-releases/main/installer.sh | sh

On macOS with Homebrew:

brew install opentelemetry-collector

3. Source Build¶

For developers, build from source using Go:

go get github.com/open-telemetry/opentelemetry-collector
cd $GOPATH/src/github.com/open-telemetry/opentelemetry-collector
make build

Run the binary with a configuration file:

./otelcol --config config.yaml

Configuration¶

The Collector uses a YAML configuration file (config.yaml) to define pipelines, exporters, and processors. A basic example:

service:
  pipelines:
    traces:
      receivers: [otlp]
      processors: [batch]
      exporters: [prometheusremotewrite, jaeger]
    metrics:
      receivers: [otlp]
      exporters: [prometheusremotewrite]

Key Configuration Elements: - Receivers: Define how the Collector accepts data (e.g., otlp, prometheus). - Processors: Transform or filter data (e.g., batch, resourcedetection). - Exporters: Specify where to send data (e.g., prometheusremotewrite, otlp).

Configuration File Location: - Linux/macOS: /etc/otelcol/ or ~/.config/otelcol/ - Windows: C:\Users\<User>\.config\otelcol\

Validation: Test your config with:

otelcol configcheck --config config.yaml

Platform-Specific Notes¶

Platform	Notes
Linux	Use `systemd` or `init.d` to manage the service. Example systemd unit file:
	`[Unit]` `Description=OpenTelemetry Collector` `[Service]` `ExecStart=/usr/local/bin/otelcol --config /etc/otelcol/config.yaml`
macOS	Homebrew installs the binary to `/usr/local/bin/otelcol`.
Windows	MSI install places binaries in `C:\Program Files\OpenTelemetry Collector`.
Cloud	Use container registries (e.g., Docker Hub) or serverless functions.

Diagram: Collector Architecture¶

graph TD
    A[Application] --> B[OTLP Exporter]
    B --> C[OpenTelemetry Collector]
    C --> D1[Prometheus Remote Write]
    C --> D2[Jaeger]
    C --> D3[OTLP Endpoint]
    style C fill:#4CAF50,stroke:#388E3C

Key takeaways¶

Installation: Choose Docker, package managers, or source builds based on your environment.
Configuration: Define receivers, processors, and exporters in config.yaml to tailor data flow.
Validation: Use otelcol configcheck to ensure your config is syntactically correct.
Platform Adaptation: Adjust installation steps for Linux, macOS, Windows, or cloud environments.