Local environments and CI

Workspace environments are standard conda prefixes stored under the workspace root. They support normal conda activation and execution workflows.

Workspace environments

All environments are installed under .conda/envs/ in the workspace root:

my-project/
├── conda.toml
├── conda.lock
├── .conda/
│   └── envs/
│       ├── default/
│       ├── test/
│       └── docs/
└── src/

Environments are standard conda prefixes. Use conda workspace run -e <name> -- CMD to run a command in an environment, conda workspace shell -e <name> for an interactive shell, or conda activate .conda/envs/<name> directly.

Optimizing disk usage with hardlinks

conda hardlinks packages from its global cache into environment prefixes to save disk space. In CI and Docker the global cache is often on a different filesystem or volume from the workspace directory, causing conda to silently fall back to copying packages. That roughly doubles disk usage per environment.

Set the CONDA_PKGS_DIRS environment variable to a path inside the workspace before installing so that the cache and environments share a filesystem:

export CONDA_PKGS_DIRS="$PWD/.conda/pkgs"
conda workspace install

GitHub Actions

- name: Install workspace
  run: |
    export CONDA_PKGS_DIRS="$PWD/.conda/pkgs"
    conda workspace install

GitLab CI

install:
  script:
    - export CONDA_PKGS_DIRS="$PWD/.conda/pkgs"
    - conda workspace install

On developer workstations the global cache is usually on the same filesystem and hardlinks work without any extra configuration.