Skip to main content

Run Docker in Docker on RunPod CPU Instances

note

This tutorial applies only to RunPod's CPU offering. It is not possible to run Docker inside of a GPU at this time.

For information on building Docker images on GPUs, see Build Docker Images with Bazel.

RunPod provides virtualized machine images on CPUs. With this virtualization, you can use your own Docker image as a base image and, in addition, run Docker containers inside the CPU instance.

Create a CPU Instance​

To get started, create a CPU instance:

  1. Navigate to Pods and select + Deploy.
  2. Select your CPU type.
  3. Customize your instance by setting up the following:
    1. (recommended) Specify a Network volume.
    2. Select an instance type. For example, 32 vCPU with 256 GB of RAM.
    3. (optional) Provide a template. For example, runpod/base:0.5.1-cpu.
  4. Review your configuration and select Deploy On-Demand.

For more information, see Manage Pods.

Install Docker​

You can install Docker using one of the following methods. Ensure you have root privileges.

From the Docker Install Script​

Download and run the official Docker install script:

curl -fsSL https://get.docker.com -o get-docker.sh
sh get-docker.sh

From the Test Docker Install Script​

Download and run the Docker test install script:

curl -fsSL https://test.docker.com -o test-docker.sh
sh test-docker.sh

From the Source Repository​

Clone and install Docker from the source repository:

git clone https://github.com/docker/docker-ce.git
cd docker-ce
sh install.sh

Verify Docker Installation​

After installation, verify that Docker is installed correctly:

docker --version

Start Docker​

Start the Docker daemon:

dockerd &

Test Docker Service​

Run a test container to ensure the Docker service is running correctly:

docker run hello-world

If the container runs successfully and outputs a welcome message from Docker, you have successfully set up Docker in Docker on a RunPod CPU instance.

You now have Docker running inside a Docker container on a RunPod CPU instance. This setup allows you to manage and run additional containers as needed, leveraging the flexibility and power of Docker within a virtualized environment provided by RunPod.