Miniconda¶
![]()
Category: Development
Type: Software Installer
Tags: miniconda · python
Overview¶
The Miniconda plugin provides a lightweight, self-contained Python environment management solution for project workstations. It installs Miniconda to a shared persistent volume, creates a named conda environment with a specified Python version, and optionally installs additional packages. All workstations in the project that mount the shared volume can immediately use the managed conda environment without any additional setup.
How It Works¶
Software Installer — When added to a project, Terra downloads Miniconda, installs it to the shared volume you choose, and creates the configured conda environment with the packages you specify. The environment is then available to all workstations in the project that mount the volume.
Prerequisites¶
- A shared persistent volume provisioned in your project
- Outbound internet access from the cluster (to download the Miniconda installer and conda packages)
Installation¶
- Open Terra and navigate to the Plugin Marketplace
- Search for "Miniconda"
- Click Install
- Fill in the configuration fields below
- Click Confirm to deploy
Configuration¶
Install-Time Fields¶
| Field | Details |
|---|---|
install_volume |
shared-volume · Required Shared persistent volume to install Miniconda into |
destination |
string · Optional · Default: /minicondaDirectory path within the volume for the Miniconda installation |
environment_name |
string · Required · Default: baseName of the conda environment to create |
python_version |
string · Required · Default: 3.12Python version to use in the conda environment (e.g. 3.10, 3.11, 3.12) |
packages |
string · Optional · Default: ""Space-separated list of additional conda/pip packages to install (e.g. numpy pandas scikit-learn) |
source_url |
string · Optional Custom URL to the Miniconda installer script (overrides the default download source) |
Notes¶
- Miniconda is a minimal installer for conda — it includes only conda, Python, and a small set of useful packages. Additional packages are installed via the
packagesfield or within workstations usingconda install - The
source_urlfield is useful for air-gapped environments that mirror installer scripts internally - To use the conda environment from workstations, activate it with
conda activate <environment_name>or reference the full path at<destination>/envs/<environment_name>/bin/python