マシンエージェント Docker イメージへのアクセス

Splunk AppDynamics は、2 セットのマシンエージェント Docker イメージを公開しています。1 つはマシンエージェント用で、もう 1 つは Kubernetes 向け Analytics 対応マシンエージェント用です。Docker イメージの各セットには、非ルートおよびルートユーザー用の Debian および Alpine イメージが含まれています。

Docker Deployment for Combined Agent

Starting from version 26.3.0, all Machine Agent Docker images include the bundled Splunk OTel Collector. A built-in wrapper script serves as the container entrypoint to initialize both components. Once launched, the Machine Agent and the OTel Collector operate as independent processes, allowing you to choose and scale your monitoring strategy through environment variables. Mode selection and all configuration is done through environment variables passed with docker run -e. For more see information, see the Environment Variables section in Combined Agent for Infrastructure Visibility.

Deployment Modes

You can deploy the agent by using any one of the following modes depending on your monitoring requirements:

Machine Agent only (default)
Dual Signal
OTel only

Machine Agent Only (Default)

Use the following commands for standard infrastructure monitoring without OpenTelemetry.

CODE

docker run -d \
  -e APPDYNAMICS_AGENT_ACCOUNT_NAME=<account> \
  -e APPDYNAMICS_AGENT_ACCOUNT_ACCESS_KEY=<access-key> \
  -e APPDYNAMICS_CONTROLLER_HOST_NAME=<controller-host> \
  -e APPDYNAMICS_CONTROLLER_PORT=<port> \
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \
  appdynamics/machine-agent:<tag>

docker run -d \
  -e APPDYNAMICS_AGENT_ACCOUNT_NAME=<account> \
  -e APPDYNAMICS_AGENT_ACCOUNT_ACCESS_KEY=<access-key> \
  -e APPDYNAMICS_CONTROLLER_HOST_NAME=<controller-host> \
  -e APPDYNAMICS_CONTROLLER_PORT=<port> \
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \
  appdynamics/machine-agent:<tag>

After you run the commands, the script starts only the Machine Agent. It collects infrastructure metrics (CPU, Memory, Disk) and sends them to the AppDynamics Controller.

Dual Signal Mode

Use the following commands to send data to both the AppDynamics Controller and Splunk Observability Cloud.

CODE

docker run -d \
  -e APPDYNAMICS_AGENT_ACCOUNT_NAME=<account> \
  -e APPDYNAMICS_AGENT_ACCOUNT_ACCESS_KEY=<access-key> \
  -e APPDYNAMICS_CONTROLLER_HOST_NAME=<controller-host> \
  -e APPDYNAMICS_CONTROLLER_PORT=<port> \
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \
  -e SPLUNK_OTEL_ENABLED=true \
  -e SPLUNK_ACCESS_TOKEN=<splunk-token> \
  -e SPLUNK_REALM=<realm> \
  appdynamics/machine-agent:<tag>

docker run -d \
  -e APPDYNAMICS_AGENT_ACCOUNT_NAME=<account> \
  -e APPDYNAMICS_AGENT_ACCOUNT_ACCESS_KEY=<access-key> \
  -e APPDYNAMICS_CONTROLLER_HOST_NAME=<controller-host> \
  -e APPDYNAMICS_CONTROLLER_PORT=<port> \
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \
  -e SPLUNK_OTEL_ENABLED=true \
  -e SPLUNK_ACCESS_TOKEN=<splunk-token> \
  -e SPLUNK_REALM=<realm> \
  appdynamics/machine-agent:<tag>

After you run the commands, the script initializes two processes: the Machine Agent and the Splunk OTel Collector. The Machine Agent sends metrics to AppDynamics, while the OTel Collector streams telemetry to Splunk Observability Cloud. This provides a unified view of infrastructure and distributed tracing across both platforms.

OpenTelemetry Mode

Use the following commands to run only the Splunk OTel Collector within the container.

CODE

docker run -d \
  -e SPLUNK_OTEL_ONLY=true \
  -e SPLUNK_ACCESS_TOKEN=<splunk-token> \
  -e SPLUNK_REALM=<realm> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \          # Recommended
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \             # Optional: business context
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \                   # Optional: business context
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \                   # Optional: business context
  appdynamics/machine-agent:<tag>

docker run -d \
  -e SPLUNK_OTEL_ONLY=true \
  -e SPLUNK_ACCESS_TOKEN=<splunk-token> \
  -e SPLUNK_REALM=<realm> \
  -e APPDYNAMICS_AGENT_UNIQUE_HOST_ID=<host-id> \          # Recommended
  -e APPDYNAMICS_AGENT_APPLICATION_NAME=<app> \             # Optional: business context
  -e APPDYNAMICS_AGENT_TIER_NAME=<tier> \                   # Optional: business context
  -e APPDYNAMICS_AGENT_NODE_NAME=<node> \                   # Optional: business context
  appdynamics/machine-agent:<tag>

After you run the commands, the script disables the Machine Agent and only starts the Splunk OTel Collector. All data is sent exclusively to the Splunk Observability Cloud.

Important: Always set APPDYNAMICS_AGENT_UNIQUE_HOST_ID to a meaningful value.If this is not set, the container ID is used as the host name. This causes data fragmentation in your dashboards every time the container restarts. Using a unique ID ensures consistent historical data in both AppDynamics and Splunk.

Custom OTel Pipeline

To use a custom OpenTelemetry configuration, mount your YAML file to the container:

CODE

docker run -d \
  -v /path/to/custom_config.yaml:/opt/appdynamics/conf/otel/agent_config.yaml \
  ... (other variables) ...
  appdynamics/machine-agent:<tag>

docker run -d \
  -v /path/to/custom_config.yaml:/opt/appdynamics/conf/otel/agent_config.yaml \
  ... (other variables) ...
  appdynamics/machine-agent:<tag>

Logs

The logs logs are located at:


Component	Path (Inside Container)
Machine Agent	`/opt/appdynamics/logs/machine-agent.log`
OTel Collector	`/opt/appdynamics/logs/otel/otel-collector.log`