Ubuntu Vagrant VM for Dev

This is a headless Ubuntu 22.04 VM provisioned with:

• 8 GB RAM, 2 CPUs, 100 GB dynamic disk
• Git, Vim, Byobu, Fish shell
• Miniforge Conda (base environment)
• Byobu auto-start for SSH sessions
• Podman + podman-compose for container workflows
• Fully compatible with Dockerfiles and docker-compose.yml

---

Setup

1. Install dependencies on your Mac:

   brew install --cask virtualbox
   brew install vagrant

2. Clone or create this project folder and place the Vagrantfile inside:

   git clone https://github.com/cehbrecht/dev-machine.git
   cd dev-machine/

3. Bring up the VM:

   vagrant up
   vagrant ssh

• Byobu will auto-start.
• Bash is the default shell; Fish is available (fish).
• Conda base environment is active.

---

Using Conda

conda info          # check Conda
conda list          # installed packages
conda create -n dev python=3.11 numpy pandas tqdm
conda activate dev

---

Using Podman + podman-compose

• Build images from Dockerfiles:

  podman build -t myimage .

• Run docker-compose.yml using Podman:

  podman-compose up

• Manage containers as usual (podman ps, podman stop, etc.)

---

VS Code Remote-SSH Workflow

1. Install the Remote - SSH extension in VS Code.

2. Add a new SSH host in ~/.ssh/config:

   Host ubunti HostName 127.0.0.1 User vagrant Port 2222 IdentityFile ~/.vagrant.d/insecure_private_key

3. In VS Code, choose Remote-SSH → Connect to Host → ubunti.

4. You can now edit code locally while running inside the VM. Use /vagrant to access your synced folder:

   cd /vagrant

---

Disk

• Root filesystem: ~100 GB dynamic disk
• Plenty of space for Conda envs, datasets, and containers

---

Tips

• Shut down VM:

  vagrant halt

• Restart:

  vagrant up

• Destroy (reset to clean VM):

  vagrant destroy -f