Collector deployment

Deployment Options

The Dynatrace Collector can be deployed for the following platforms:

Kubernetes
Docker
Windows, macOS, and Linux
Linux installer packages

Kubernetes

For Kubernetes, the Dynatrace Collector can be deployed in the following ways:

OpenTelemetry Kubernetes Operator
Helm
Raw manifest

Dynatrace access details

Before you deploy the Collector, you need to set up the necessary Kubernetes secrets for the Dynatrace access details.

Use kubectl to create Kuberenetes secrets for the Dynatrace export details. Replace the placeholders (indicated by curly brackets) with the actual values for the export URL and the API token.

kubectl create secret generic dynatrace-otelcol-dt-api-credentials --from-literal=DT_ENDPOINT={ENDPOINT_URL_HERE} --from-literal=DT_API_TOKEN={API_TOKEN_HERE}

Kubernetes deployment types

The following sample configurations apply a resource limit of 512 megabytes. You may need to adjust this under resources.limits.memory for your particular use case.

Prerequisites

If you haven't installed OpenTelemetry Operator yet, first make sure cert-manager is installed. Afterwards, you can deploy Operator with the following kubectl command:

kubectl apply -f https://github.com/open-telemetry/opentelemetry-operator/releases/latest/download/opentelemetry-operator.yaml

After the installation, deploy the Dynatrace Collector either as a gateway or an agent, with one of the following configuration samples. Save it as crd-dynatrace-collector.yaml and deploy it with kubectl apply.

Deploy as a gateway (Deployment)

apiVersion: opentelemetry.io/v1beta1
kind: OpenTelemetryCollector
metadata:
  name: dynatrace-otel
spec:
  envFrom:
  - secretRef:
      name: dynatrace-otelcol-dt-api-credentials
  env:
  - name: MY_POD_IP
    valueFrom:
      fieldRef:
        fieldPath: status.podIP
  mode: "deployment"
  image: "ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest"
  resources:
    limits:
      memory: 512Mi
  config:
    receivers:
      otlp:
        protocols:
          grpc:
            endpoint: ${env:MY_POD_IP}:4317
          http:
            endpoint: ${env:MY_POD_IP}:4318

    exporters:
      otlphttp:
        endpoint: "${env:DT_ENDPOINT}"
        headers:
          Authorization: "Api-Token ${env:DT_API_TOKEN}"

    service:
      pipelines:
        traces:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        metrics:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        logs:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]

Deploy as an agent (DaemonSet)

apiVersion: opentelemetry.io/v1beta1
kind: OpenTelemetryCollector
metadata:
  name: dynatrace-otel
spec:
  envFrom:
  - secretRef:
      name: dynatrace-otelcol-dt-api-credentials
  env:
  - name: MY_POD_IP
    valueFrom:
      fieldRef:
        fieldPath: status.podIP
  mode: "daemonset"
  image: "ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest"
  resources:
    limits:
      memory: 512Mi
  config:
    receivers:
      otlp:
        protocols:
          grpc:
            endpoint: ${env:MY_POD_IP}:4317
          http:
            endpoint: ${env:MY_POD_IP}:4318

    exporters:
      otlphttp:
        endpoint: "${env:DT_ENDPOINT}"
        headers:
          Authorization: "Api-Token ${env:DT_API_TOKEN}"

    service:
      pipelines:
        traces:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        metrics:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        logs:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]

Choose one of the following deployment options.

The Helm charts below use alternateConfig to provide the Collector configuration. With this entry, the default Helm chart configuration, as well as a possibly present config object, will be ignored.

Deploy as a gateway (Deployment)

Save the following YAML configuration under values-deployment.yaml

mode: deployment
image:
  repository: ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector
  tag: latest
command:
  name: dynatrace-otel-collector
extraEnvs:
- name: DT_API_TOKEN
  valueFrom:
    secretKeyRef:
      name: dynatrace-otelcol-dt-api-credentials
      key: DT_API_TOKEN
- name: DT_ENDPOINT
  valueFrom:
    secretKeyRef:
      name: dynatrace-otelcol-dt-api-credentials
      key: DT_ENDPOINT
resources:
  limits:
    memory: 512Mi
alternateConfig:
  receivers:
    otlp:
      protocols:
        grpc:
          endpoint: ${env:MY_POD_IP}:4317
        http:
          endpoint: ${env:MY_POD_IP}:4318
    filelog: null
  exporters:
    otlphttp:
      endpoint: "${env:DT_ENDPOINT}"
      headers:
        Authorization: "Api-Token ${env:DT_API_TOKEN}"
  service:
    pipelines:
      traces:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]
      metrics:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]
      logs:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]

Run the following commands to configure and install the Helm charts

helm repo add open-telemetry https://open-telemetry.github.io/opentelemetry-helm-charts
helm repo update
helm upgrade -i dynatrace-collector open-telemetry/opentelemetry-collector -f values-deployment.yaml

Deploy as an agent (DaemonSet)

Save the following YAML configuration under values-daemonset.yaml

mode: daemonset
image:
  repository: ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector
  tag: latest
command:
  name: dynatrace-otel-collector
extraEnvs:
- name: DT_API_TOKEN
  valueFrom:
    secretKeyRef:
      name: dynatrace-otelcol-dt-api-credentials
      key: DT_API_TOKEN
- name: DT_ENDPOINT
  valueFrom:
    secretKeyRef:
      name: dynatrace-otelcol-dt-api-credentials
      key: DT_ENDPOINT
resources:
  limits:
    memory: 512Mi
alternateConfig:
  receivers:
    otlp:
      protocols:
        grpc:
          endpoint: ${env:MY_POD_IP}:4317
        http:
          endpoint: ${env:MY_POD_IP}:4318

  exporters:
    otlphttp:
      endpoint: "${env:DT_ENDPOINT}"
      headers:
        Authorization: "Api-Token ${env:DT_API_TOKEN}"

  service:
    pipelines:
      traces:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]
      metrics:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]
      logs:
        receivers: [otlp]
        processors: []
        exporters: [otlphttp]

Run the following commands to configure and install the Helm charts

helm repo add open-telemetry https://open-telemetry.github.io/opentelemetry-helm-charts
helm repo update
helm upgrade -i dynatrace-collector open-telemetry/opentelemetry-collector -f values-daemonset.yaml

Network ports

Make sure to configure and forward all relevant network ports, using the ports configuration value.

Use kubectl apply with the following configuration to set up the service definitions and configure the correct ports.

apiVersion: v1
kind: Service
metadata:
  name: dynatrace-otel-collector
spec:
  internalTrafficPolicy: Cluster
  ipFamilies:
  - IPv4
  ipFamilyPolicy: SingleStack
  ports:
  - name: jaeger-compact
    port: 6831
    protocol: UDP
    targetPort: 6831
  - name: jaeger-grpc
    port: 14250
    protocol: TCP
    targetPort: 14250
  - name: jaeger-thrift
    port: 14268
    protocol: TCP
    targetPort: 14268
  - appProtocol: grpc
    name: otlp
    port: 4317
    protocol: TCP
    targetPort: 4317
  - name: otlp-http
    port: 4318
    protocol: TCP
    targetPort: 4318
  - name: zipkin
    port: 9411
    protocol: TCP
    targetPort: 9411
  selector:
    app.kubernetes.io/instance: dynatrace-otel-collector
    app.kubernetes.io/name: dynatrace-otel-collector
  sessionAffinity: None
  type: ClusterIP

Create a ConfigMap by applying the following configuration with kubectl apply to set up the Collector configuration.

apiVersion: v1
kind: ConfigMap
metadata:
  name: dynatrace-otel-collector-config
data:
  otel-collector-config: |
    receivers:
      otlp:
        protocols:
          grpc:
            endpoint: ${env:MY_POD_IP}:4317
          http:
            endpoint: ${env:MY_POD_IP}:4318
    exporters:
      otlphttp:
        endpoint: "${env:DT_ENDPOINT}"
        headers:
          Authorization: "Api-Token ${env:DT_API_TOKEN}"
    service:
      pipelines:
        traces:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        metrics:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]
        logs:
          receivers: [otlp]
          processors: []
          exporters: [otlphttp]

Apply the following manifest configuration with kubectl apply to create the Dynatrace Collector Deployment.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: dynatrace-otel-collector
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: dynatrace-otel-collector
  replicas: 1
  template:
    metadata:
      labels:
        app.kubernetes.io/instance: dynatrace-otel-collector
        app.kubernetes.io/name: dynatrace-otel-collector
    spec:
      #You may have to configure RBAC to grant proper permissions for enriching data
      #serviceAccountName: dynatrace-otel-collector
      containers:
        - args: ["--config", "/conf/otel-collector-config.yaml"]
          env:
            - name: MY_POD_IP
              valueFrom:
                fieldRef:
                  apiVersion: v1
                  fieldPath: status.podIP
            - name: DT_ENDPOINT
              valueFrom:
                secretKeyRef:
                  name: dynatrace-otelcol-dt-api-credentials
                  key: DT_ENDPOINT
            - name: DT_API_TOKEN
              valueFrom:
                secretKeyRef:
                  name: dynatrace-otelcol-dt-api-credentials
                  key: DT_API_TOKEN
          image: ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest
          name: otel-collector
          resources:
            limits:
              memory: 512Mi
          ports:
            - containerPort: 8888 # Default endpoint for querying metrics of prometheus exporter.
          volumeMounts:
            - name: dynatrace-otel-collector-config
              mountPath: /conf
      volumes:
        - configMap:
            name: dynatrace-otel-collector-config
            items:
              - key: otel-collector-config
                path: otel-collector-config.yaml
          name: dynatrace-otel-collector-config

Service account

In Kubernetes, it is common to enrich OpenTelemetry signals using the Kubernetes Attributes processor. This requires a Kubernetes service account, which is automatically configured when using Operator or Helm.

For raw manifests, this needs to be configured manually by adding a spec.serviceAccountName: collector entry to the deployment manifest.

Docker

Run the following command to download the image of the Dynatrace Collector:

docker pull ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest

Next, run the Collector image with the following command:

docker run -v $(pwd)/otel-collector-config.yaml:/etc/otelcol/otel-collector-config.yaml ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest --config=/etc/otelcol/otel-collector-config.yaml

The -v parameter maps the local configuration file to the given container path, which is subsequently passed to the --config parameter.

Make sure to map all required network ports with the -p parameter. For example, if you accept OTLP gRPC requests on the default port, you need to specify port 4317. For OTLP over HTTP specify port 4318.

Docker Compose

Use the following configuration in your Compose file to deploy and run the Collector image:

version: "3"
services:
  collector:
    image: ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest
    command: ["--config=/etc/otelcol/otel-collector-config.yaml"]
    volumes:
      - ./otel-collector-config.yaml:/etc/otelcol/otel-collector-config.yaml
    ports:
      - "4317:4317"   # OTLP gRPC
      - "4318:4318"   # OTLP HTTP

In the example above, ports is configured for gRPC and HTTP. Adjust the list of ports according to your specific use case.

Windows, macOS, and Linux

To install the Collector binary manually

Download the latest version for your operating system from GitHub.
Decompress the archive file.
Set up the desired configuration and save it to otel-collector-config.yaml.
Run the dynatrace-otel-collector binary and pass the path to the configuration file using the --config parameter.
```
./dynatrace-otel-collector --config=$(pwd)/otel-collector-config.yaml
```

Linux installer packages

Dynatrace also provides DEB and RPM installer packages for Linux systems on x86-64 and ARM64 architectures.

Required init system

The installer packages require Systemd to be the active init system.

To deploy the Collector using an installer package, download the latest version for your operating system from GitHub, and install it using root privileges and the following commands.

Replace the following two placeholders in the commands with their actual content:

<VERSION> with the version tag of the download
<ARCH> with the system architecture tag (that is, x86_64 or arm64) of the download

apt-get update
dpkg -i dynatrace-otel-collector_<VERSION>_Linux_<ARCH>.deb

yum update
rpm -ivh dynatrace-otel-collector_<VERSION>_Linux_<ARCH>.rpm

Service configuration

When first starting the service, it may fail to start if there is no configuration file in place yet. By default, the Collector attempts to find the file at /etc/dynatrace-otel-collector/config.yaml.

Custom configuration location

If you wish to use a different path, you can override the default path with the --config parameter as part of the OTELCOL_OPTIONS variable in the Systemd environment file at /etc/dynatrace-otel-collector/dynatrace-otel-collector.conf:

OTELCOL_OPTIONS="--config=<HERE-PATH-TO-CONFIG-FILE>"

Subsequent package updates will replace this file, so be sure to back up and restore its content during an update of the Collector package. Alternatively, you can override the configuration with the systemctl edit command.

To see all available configuration options, run the Collector binary with the --help parameter.

After changing the configuration, make sure to restart the service using the following command and root privileges:

systemctl restart dynatrace-otel-collector

Service status

To view the current status of the Collector service, run the following command with root privileges:

systemctl status dynatrace-otel-collector

To check the output of the Collector service, run the following command with root privileges:

journalctl -u dynatrace-otel-collector

Container image registries

Container images for the Dynatrace distribution of the OpenTelemetry Collector are available in

GitHub Container Registry (GHCR)
- ghcr.io/dynatrace/dynatrace-otel-collector/dynatrace-otel-collector:latest
Amazon Elastic Container Registry (Amazon ECR)
- public.ecr.aws/dynatrace/dynatrace-otel-collector:latest

Deployment Modes

The Collector can be deployed as a standalone agent, a gateway, or a sidecar.

As an agent, the Collector is deployed either with the application or (recommended) on the same host as the application. This Collector can receive telemetry data and enhance it with, for example, tags or infrastructure information.

OpenTelemetry collector as agent

As a gateway, one or more Collector instances can be deployed as standalone services. This Collector can be deployed additionally, for example, per cluster, region, or data center. A load balancer can help scale the independently operating Collector instances.

OpenTelemetry collector as gateway

In Kubernetes clusters, you can use a variant of the agent mode and deploy the Collector as a sidecar to your application pods.

OpenTelemetry collector as sidecar

You need to use the OpenTelemetry operator for this deployment option. Be aware that using the Collector as a sidecar might increase resource consumption.

It's also possible to combine these deployment modes and chain Collector instances. Consider this if you are deploying the Collector in large environments.