Docker Ros2 Development

Docker-Based ROS2 Development Skill

When to Use This Skill

• Writing Dockerfiles for ROS2 workspaces with colcon builds
• Setting up docker-compose for multi-container robotic systems
• Debugging DDS discovery failures between containers (CycloneDDS, FastDDS)
• Configuring GPU passthrough with NVIDIA Container Toolkit for perception nodes
• Forwarding X11 or Wayland displays for rviz2 and rqt tools
• Managing USB device passthrough for cameras, LiDARs, and serial devices
• Building CI/CD pipelines with Docker-based ROS2 builds and test runners
• Creating devcontainer configurations for VS Code with ROS2 extensions
• Optimizing Docker layer caching for colcon workspace builds
• Designing dev-vs-deploy container strategies with multi-stage builds

ROS2 Docker Image Hierarchy

Official OSRF images follow a layered hierarchy. Always choose the smallest base that satisfies dependencies.

┌──────────────────────────────────────────────────────────────────┐
│  ros:-desktop-full  (~3.5 GB)                            │
│  ┌────────────────────────────────────────────────────────────┐  │
│  │  ros:-desktop     (~2.8 GB)                        │  │
│  │  ┌──────────────────────────────────────────────────────┐  │  │
│  │  │  ros:-perception (~2.2 GB)                    │  │  │
│  │  │  ┌────────────────────────────────────────────────┐   │  │  │
│  │  │  │  ros:-ros-base  (~1.1 GB)              │   │  │  │
│  │  │  │  ┌──────────────────────────────────────────┐  │   │  │  │
│  │  │  │  │  ros:-ros-core (~700 MB)         │  │   │  │  │
│  │  │  │  └──────────────────────────────────────────┘  │   │  │  │
│  │  │  └────────────────────────────────────────────────┘   │  │  │
│  │  └──────────────────────────────────────────────────────┘  │  │
│  └────────────────────────────────────────────────────────────┘  │
└──────────────────────────────────────────────────────────────────┘

| Image Tag | Base OS | Size | Contents | Use Case | |--------------------------|----------------|---------|---------------------------------------------|-------------------------------------| | ros:humble-ros-core | Ubuntu 22.04 | ~700 MB | rclcpp, rclpy, rosout, launch | Minimal runtime for single nodes | | ros:humble-ros-base | Ubuntu 22.04 | ~1.1 GB | ros-core + commoninterfaces, rosbag2 | Most production deployments | | ros:humble-perception | Ubuntu 22.04 | ~2.2 GB | ros-base + imagetransport, cvbridge, PCL | Camera/lidar perception pipelines | | ros:humble-desktop | Ubuntu 22.04 | ~2.8 GB | perception + rviz2, rqt, demos | Development with GUI tools | | ros:jazzy-ros-core | Ubuntu 24.04 | ~750 MB | rclcpp, rclpy, rosout, launch | Minimal runtime (Jazzy/Noble) | | ros:jazzy-ros-base | Ubuntu 24.04 | ~1.2 GB | ros-core + commoninterfaces, rosbag2 | Production deployments (Jazzy) |

Multi-Stage Dockerfiles for ROS2

Dev Stage

The development stage includes build tools, debuggers, and editor support for interactive use.

FROM ros:humble-desktop AS dev
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential cmake gdb python3-pip \
    python3-colcon-common-extensions python3-rosdep \
    ros-humble-ament-lint-auto ros-humble-ament-cmake-pytest \
    ccache \
    && rm -rf /var/lib/apt/lists/*
ENV CCACHE_DIR=/ccache
ENV CC="ccache gcc"
ENV CXX="ccache g++"

Build Stage

Copies only src/ and package.xml files to maximize cache hits during dependency resolution.

FROM ros:humble-ros-base AS build
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3-colcon-common-extensions python3-rosdep \
    && rm -rf /var/lib/apt/lists/*
WORKDIR /ros2_ws
# Copy package manifests first for dependency caching
COPY src/my_pkg/package.xml src/my_pkg/package.xml
RUN . /opt/ros/humble/setup.sh && apt-get update && \
    rosdep install --from-paths src --ignore-src -r -y && \
    rm -rf /var/lib/apt/lists/*
# Source changes invalidate only this layer and below
COPY src/ src/
RUN . /opt/ros/humble/setup.sh && \
    colcon build --cmake-args -DCMAKE_BUILD_TYPE=Release \
      --event-handlers console_direct+

Runtime Stage

Contains only the built install space and runtime dependencies. No compilers, no source code.

FROM ros:humble-ros-core AS runtime
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3-yaml ros-humble-rmw-cyclonedds-cpp \
    && rm -rf /var/lib/apt/lists/*
COPY --from=build /ros2_ws/install /ros2_ws/install
RUN groupadd -r rosuser && useradd -r -g rosuser -m rosuser
USER rosuser
COPY ros_entrypoint.sh /ros_entrypoint.sh
ENTRYPOINT ["/ros_entrypoint.sh"]
CMD ["ros2", "launch", "my_pkg", "bringup.launch.py"]

Full Multi-Stage Dockerfile

# syntax=docker/dockerfile:1
# Usage:
#   docker build --target dev -t my_robot:dev .
#   docker build --target runtime -t my_robot:latest .
ARG ROS_DISTRO=humble
ARG BASE_IMAGE=ros:${ROS_DISTRO}-ros-base

# Stage 1: Dependency base — install apt and rosdep deps
FROM ${BASE_IMAGE} AS deps
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3-colcon-common-extensions python3-rosdep \
    && rm -rf /var/lib/apt/lists/*
WORKDIR /ros2_ws
# Copy only package.xml files for rosdep resolution (maximizes cache reuse)
COPY src/my_robot_bringup/package.xml src/my_robot_bringup/package.xml
COPY src/my_robot_perception/package.xml src/my_robot_perception/package.xml
COPY src/my_robot_msgs/package.xml src/my_robot_msgs/package.xml
COPY src/my_robot_navigation/package.xml src/my_robot_navigation/package.xml
RUN . /opt/ros/${ROS_DISTRO}/setup.sh && \
    apt-get update && \
    rosdep install --from-paths src --ignore-src -r -y && \
    rm -rf /var/lib/apt/lists/*

# Stage 2: Development — full dev environment
FROM deps AS dev
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential gdb valgrind ccache python3-pip python3-pytest \
    ros-${ROS_DISTRO}-ament-lint-auto \
    ros-${ROS_DISTRO}-launch-testing-ament-cmake \
    ros-${ROS_DISTRO}-rviz2 ros-${ROS_DISTRO}-rqt-graph \
    && rm -rf /var/lib/apt/lists/*
ENV CCACHE_DIR=/ccache CC="ccache gcc" CXX="ccache g++"
COPY src/ src/
COPY ros_entrypoint.sh /ros_entrypoint.sh
ENTRYPOINT ["/ros_entrypoint.sh"]
CMD ["bash"]

# Stage 3: Build — compile workspace
FROM deps AS build
COPY src/ src/
RUN . /opt/ros/${ROS_DISTRO}/setup.sh && \
    colcon build \
      --cmake-args -DCMAKE_BUILD_TYPE=Release -DBUILD_TESTING=OFF \
      --event-handlers console_direct+ \
      --parallel-workers $(nproc)

# Stage 4: Runtime — minimal production image
FROM ros:${ROS_DISTRO}-ros-core AS runtime
ARG ROS_DISTRO=humble
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3-yaml ros-${ROS_DISTRO}-rmw-cyclonedds-cpp \
    && rm -rf /var/lib/apt/lists/*
COPY --from=build /ros2_ws/install /ros2_ws/install
RUN groupadd -r rosuser && useradd -r -g rosuser -m -s /bin/bash rosuser
USER rosuser
ENV RMW_IMPLEMENTATION=rmw_cyclonedds_cpp
COPY ros_entrypoint.sh /ros_entrypoint.sh
ENTRYPOINT ["/ros_entrypoint.sh"]
CMD ["ros2", "launch", "my_robot_bringup", "robot.launch.py"]

The entrypoint script both dev and runtime stages use:

#!/bin/bash
set -e
source /opt/ros/${ROS_DISTRO}/setup.bash
if [ -f /ros2_ws/install/setup.bash ]; then
    source /ros2_ws/install/setup.bash
fi
exec "$@"

Docker Compose for Multi-Container ROS2 Systems

Basic Multi-Container Setup

Each ROS2 subsystem runs in its own container with process isolation, independent scaling, and per-service resource limits.

# docker-compose.yml
version: "3.8"

x-ros-common: &ros-common
  environment:
    - ROS_DOMAIN_ID=${ROS_DOMAIN_ID:-0}
    - RMW_IMPLEMENTATION=rmw_cyclonedds_cpp
    - CYCLONEDDS_URI=file:///cyclonedds.xml
  volumes:
    - ./config/cyclonedds.xml:/cyclonedds.xml:ro
    - /dev/shm:/dev/shm
  network_mode: host
  restart: unless-stopped

services:
  rosbridge:
    
      ros2 launch my_robot_navigation navigation.launch.py
        use_sim_time:=false map:=/maps/warehouse.yaml
    volumes:
      - ./maps:/maps:ro

  driver:
    

  
    
      
        
      
      false
    
    
      
      
        
        
        
        
      
      auto
      120
    
    
      
    
  

FastDDS XML Config

  
    
      
        
          
            SIMPLE
            10
          
          
            
              perception7412
            
            
              navigation7412
            
            
              driver7412
            
          
        
      
    
  

Mount and activate in compose:

# CycloneDDS
environment:
  - RMW_IMPLEMENTATION=rmw_cyclonedds_cpp
  - CYCLONEDDS_URI=file:///cyclonedds.xml
volumes:
  - ./config/cyclonedds.xml:/cyclonedds.xml:ro

# FastDDS
environment:
  - RMW_IMPLEMENTATION=rmw_fastrtps_cpp
  - FASTRTPS_DEFAULT_PROFILES_FILE=/fastdds.xml
volumes:
  - ./config/fastdds.xml:/fastdds.xml:ro

Shared Memory Transport in Docker

DDS shared memory (zero-copy) requires /dev/shm sharing between containers. This provides highest throughput for large messages (images, point clouds).

services:
  perception:
    shm_size: "512m"                    # Default 64 MB is too small for image topics
    volumes:
      - /dev/shm:/dev/shm              # Share host shm for inter-container zero-copy

  
    
      true
    
  

Constraints: all communicating containers must share /dev/shm or use ipc: host. Use --ipc=shareable on one container and --ipc=container: on others for scoped sharing.

Networking Modes and ROS2 Implications

Host Networking

services:
  my_node:
    network_mode: host      # Shares host network namespace; DDS multicast works natively

Bridge Networking (Default)

services:
  my_node:
    networks: [ros_net]
networks:
  ros_net:
    driver: bridge          # DDS multicast blocked; requires unicast peer config

Macvlan Networking

networks:
  ros_macvlan:
    driver: macvlan
    driver_opts:
      parent: eth0
    ipam:
      config:
        - subnet: 192.168.1.0/24
          gateway: 192.168.1.1
services:
  my_node:
    networks:
      ros_macvlan:
        ipv4_address: 192.168.1.50   # Real LAN IP; DDS multicast works natively

Decision Table

| Factor | Host | Bridge | Macvlan | |---------------------|------------------|-------------------------|-----------------------| | DDS discovery | Works natively | Needs unicast peers | Works natively | | Network isolation | None | Full isolation | LAN-level isolation | | Port conflicts | Yes (host ports) | No (mapped ports) | No (unique IPs) | | Performance | Native | Slight overhead | Near-native | | Multi-host support | No | With overlay networks | Yes (same LAN) | | When to use | Dev, single host | CI/CD, multi-tenant | Multi-robot on LAN |

GPU Passthrough for Perception

NVIDIA Container Toolkit Setup

# Install NVIDIA Container Toolkit on the host
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey \
  | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg
curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \
  | sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' \
  | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit
sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker

Compose Config with deploy.resources

services:
  perception:
    image: my_robot_perception:latest
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1                    # Number of GPUs (or "all")
              capabilities: [gpu]
    environment:
      - NVIDIA_VISIBLE_DEVICES=all
      - NVIDIA_DRIVER_CAPABILITIES=compute,utility,video
    shm_size: "1g"                        # Large shm for GPUCPU transfers

For Dockerfiles that need CUDA, start from NVIDIA base and install ROS2 on top:

FROM nvidia/cuda:12.2.0-cudnn8-runtime-ubuntu22.04 AS perception-base
RUN apt-get update && apt-get install -y --no-install-recommends \
    curl gnupg2 lsb-release \
    && curl -sSL https://raw.githubusercontent.com/ros/rosdistro/master/ros.key \
       -o /usr/share/keyrings/ros-archive-keyring.gpg \
    && echo "deb [arch=$(dpkg --print-architecture) \
       signed-by=/usr/share/keyrings/ros-archive-keyring.gpg] \
       http://packages.ros.org/ros2/ubuntu $(lsb_release -cs) main" \
       > /etc/apt/sources.list.d/ros2.list \
    && apt-get update && apt-get install -y --no-install-recommends \
       ros-humble-ros-base ros-humble-cv-bridge ros-humble-image-transport \
    && rm -rf /var/lib/apt/lists/*

Verification

docker compose exec perception bash -c '
  nvidia-smi
  python3 -c "import torch; print(f\"CUDA available: {torch.cuda.is_available()}\")"
'

Display Forwarding

X11 Forwarding

services:
  rviz:
    image: my_robot:dev
    command: ros2 run rviz2 rviz2
    environment:
      - DISPLAY=${DISPLAY:-:0}                   # Forward host display
      - QT_X11_NO_MITSHM=1                       # Disable MIT-SHM (crashes in Docker)
    volumes:
      - /tmp/.X11-unix:/tmp/.X11-unix:rw          # X11 socket
      - ${HOME}/.Xauthority:/root/.Xauthority:ro  # Auth cookie
    network_mode: host

# Allow local Docker containers to access the X server
xhost +local:docker
# More secure variant:
xhost +SI:localuser:$(whoami)

Wayland Forwarding

services:
  rviz:
    image: my_robot:dev
    command: ros2 run rviz2 rviz2
    environment:
      - WAYLAND_DISPLAY=${WAYLAND_DISPLAY:-wayland-0}
      - XDG_RUNTIME_DIR=/run/user/1000
      - QT_QPA_PLATFORM=wayland
    volumes:
      - ${XDG_RUNTIME_DIR}/${WAYLAND_DISPLAY}:/run/user/1000/${WAYLAND_DISPLAY}:rw

Headless Rendering

For CI/CD or remote machines without a physical display:

# Run rviz2 headless with Xvfb for screenshot capture or testing
docker run --rm my_robot:dev bash -c '
  apt-get update && apt-get install -y xvfb mesa-utils &&
  Xvfb :99 -screen 0 1920x1080x24 &
  export DISPLAY=:99
  source /opt/ros/humble/setup.bash
  ros2 run rviz2 rviz2 -d /config/test.rviz --screenshot /output/frame.png
'

Volume Mounts and Workspace Overlays

Source Mounts for Dev

Mount only src/ during development. Let colcon write build/, install/, and log/ inside named volumes to avoid bind mount performance issues.

# BAD: mounting entire workspace — build artifacts on bind mount are slow
# volumes:
#   - ./my_ros2_ws:/ros2_ws

# GOOD: mount only source, use named volumes for build artifacts
services:
  dev:
    image: my_robot:dev
    volumes:
      - ./src:/ros2_ws/src:rw                     # Source code (bind mount)
      - build_vol:/ros2_ws/build                   # Build artifacts (named volume)
      - install_vol:/ros2_ws/install               # Install space (named volume)
      - log_vol:/ros2_ws/log                       # Log output (named volume)
    working_dir: /ros2_ws

volumes:
  build_vol:
  install_vol:
  log_vol:

ccache Caching

Persist ccache across container rebuilds for faster C++ compilation:

services:
  dev:
    volumes:
      - ccache_vol:/ccache
    environment:
      - CCACHE_DIR=/c

…

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [arpitg1304](https://github.com/arpitg1304)
- **Source:** [arpitg1304/robotics-agent-skills](https://github.com/arpitg1304/robotics-agent-skills)
- **License:** Apache-2.0

Install and usage instructions live in the source repository linked above.