Dockyard is a terminal-based Docker management tool built specifically for resource-constrained VPS servers.
Written in Rust with performance as a first-class concern, it provides complete container and image lifecycle management with real-time monitoring—all with minimal overhead and an easy-on-the-eyes TUI.
- Low-spec servers (1–4 vCPU, 512 MB–8 GB RAM)
- Production environments where every MB counts
- Development servers running multiple containers
- Raspberry Pi and other ARM-based systems
- Written in Rust with
async/await— no GIL, no garbage collection - Full image management — pull, remove, inspect, and filter images
- Container lifecycle controls — start, stop, restart, pause, unpause, remove
- Interactive shell access — exec into running containers with full TTY support
- Viewport-aware stats fetching — only queries containers visible on screen
- Staggered Docker API requests — avoids CPU spikes on small systems
- Synchronous UI rendering — zero async overhead, instant frame updates
- Event-driven architecture with a lock-free UI
- Built-in Wiki — access documentation and troubleshooting via
? - Configurable polling intervals — tune performance for your environment
- Rust 1.75+ (Install Rust)
- Docker daemon running
- Access to the Docker socket
# Download the latest release (v0.3.1)
wget https://github.com/905timur/dockyard/releases/download/v0.3.1/dockyard-v0.3.1-x86_64.tar.gz
# Extract and install
tar -xzf dockyard-v0.3.1-x86_64.tar.gz
sudo mv dockyard /usr/local/bin/
# Run
dockyard# Clone the repository
git clone https://github.com/905timur/dockyard.git
cd dockyard
# Build and run (release mode for best performance)
cargo run --releasecargo install --git https://github.com/905timur/dockyard.git --tag v0.3.1
dockyardEnsure your user can access the Docker socket:
sudo usermod -aG docker $USER
# Log out and log back in for changes to take effect# Run with default settings (3-second stats interval)
dockyard
# Adjust stats polling interval (1–10 seconds)
dockyard --stats-interval 5See the repo wiki page or help menu inside the application.
