Settings#

This page contains an overview of many important settings available in conda with examples where possible.

General configuration#

channels: Channel locations#

Listing channel locations in the .condarc file overrides conda defaults, causing conda to search only the channels listed there in the order given.

Use defaults to automatically include all default channels. Non-URL channels are interpreted as Anaconda.org user or organization names. You can change this by modifying the channel_alias as described in channel_alias: Set a channel alias. The default is just defaults.

Example:

channels:
  - <anaconda_dot_org_username>
  - http://some.custom/channel
  - file:///some/local/directory
  - defaults

To select channels for a single environment, put a .condarc file in the root directory of that environment (or use the --env option when using conda config).

Example: If you have installed Miniconda with Python 3 in your home directory and the environment is named "flowers", the path may be:

~/miniconda3/envs/flowers/.condarc

default_channels: Default channels#

Normally, the defaults channel points to several channels at the repo.anaconda.com repository, but if default_channels is defined, it sets the new list of default channels. This is especially useful for airgapped and enterprise installations.

To ensure that all users only pull packages from an on-premises repository, an administrator can set both channel alias and default_channels.

default_channels:
  - http://some.custom/channel
  - file:///some/local/directory

auto_update_conda: Update conda automatically#

When True, conda updates itself any time a user updates or installs a package in the root environment. When False, conda updates itself only if the user manually issues a conda update command. The default is True.

Example:

auto_update_conda: False

always_yes: Always yes#

Choose the yes option whenever asked to proceed, such as when installing. Same as using the --yes flag at the command line. The default is False.

Example:

always_yes: True

show_channel_urls: Show channel URLs#

Show channel URLs in conda list and when displaying what is going to be downloaded. The default is False.

Example:

show_channel_urls: True

changeps1: Change command prompt#

When using conda activate, change the command prompt from $PS1 to include the activated environment. The default is True.

Example:

changeps1: False

add_pip_as_python_dependency: Add pip as Python dependency#

Add pip, wheel, and setuptools as dependencies of Python. This ensures that pip, wheel, and setuptools are always installed any time Python is installed. The default is True.

Example:

add_pip_as_python_dependency: False

use_pip: Use pip#

Use pip when listing packages with conda list. This does not affect any conda command or functionality other than the output of the command conda list. The default is True.

Example:

use_pip: False

proxy_servers: Configure conda for use behind a proxy server#

By default, proxy settings are pulled from the HTTP_PROXY and HTTPS_PROXY environment variables or the system. Setting them here overrides that default:

proxy_servers:
    http: http://user:[email protected]:8080
    https: https://user:[email protected]:8080

To give a proxy for a specific scheme and host, use the scheme://hostname form for the key. This matches for any request to the given scheme and exact host name:

proxy_servers:
  'http://10.20.1.128': 'http://10.10.1.10:5323'

If you do not include the username and password or if authentication fails, conda prompts for a username and password.

If your password contains special characters, you need to escape them as described in Percent-encoding reserved characters on Wikipedia.

Be careful not to use http when you mean https or https when you mean http.

ssl_verify: SSL verification#

If you are behind a proxy that does SSL inspection, such as a Cisco IronPort Web Security Appliance (WSA), you may need to use ssl_verify to override the SSL verification settings.

By default, this variable is True, which means that SSL verification is used and conda verifies certificates for SSL connections. Setting this variable to False disables the connection's normal security and is not recommended:

ssl_verify: False

You can also set ssl_verify to a string path to a certificate, which can be used to verify SSL connections:

ssl_verify: corp.crt

offline: Offline mode only#

Filters out all channel URLs that do not use the file:// protocol. The default is False.

Example:

offline: True

Advanced configuration#

channel_alias: Set a channel alias#

Whenever you use the -c or --channel flag to give conda a channel name that is not a URL, conda prepends the channel_alias to the name that it was given. The default channel_alias is https://conda.anaconda.org.

If channel_alias is set to https://my.anaconda.repo:8080/conda/, then a user who runs the command conda install -c conda-forge some-package will install the package some-package from https://my.anaconda.repo:8080/conda/conda-forge.

For example, the command:

conda install --channel asmeurer <package>

is the same as:

conda install --channel https://conda.anaconda.org/asmeurer <package>

You can set channel_alias to your own repository.

Example: To set channel_alias to your repository at https://your.repo.com:

channel_alias: https://your.repo/

On Windows, you must include a slash ("/") at the end of the URL:

Example: https://your.repo/conda/

When channel_alias set to your repository at https://your.repo.com:

conda install --channel jsmith <package>

is the same as:

conda install --channel https://your.repo.com/jsmith <package>

create_default_packages: Always add packages by default#

When creating new environments, add the specified packages by default. The default packages are installed in every environment you create. You can override this option at the command prompt with the --no-default-packages flag. The default is to not include any packages.

Example:

create_default_packages:
  - pip
  - ipython
  - scipy=0.15.0

track_features: Track features#

Enable certain features to be tracked by default. The default is to not track any features. This is similar to adding MKL to the create_default_packages list.

Example:

track_features:
  - mkl

update_dependencies: Disable updating of dependencies#

By default, conda install updates the given package to the latest version and installs any dependencies necessary for that package. However, if dependencies that satisfy the package's requirements are already installed, conda will not update those packages to the latest version.

In this case, if you would prefer that conda update all dependencies to the latest version that is compatible with the environment, set update_dependencies to True.

The default is False.

Example:

update_dependencies: True

Note

Conda still ensures that dependency specifications are satisfied. Thus, some dependencies may still be updated or, conversely, this may prevent packages given at the command line from being updated to their latest versions. You can always specify versions at the command line to force conda to install a given version, such as conda install numpy=1.9.3.

To avoid updating only specific packages in an environment, a better option may be to pin them. For more information, see Preventing packages from updating (pinning).

disallow: Disallow installation of specific packages#

Disallow the installation of certain packages. The default is to allow installation of all packages.

Example:

disallow:
  - anaconda

add_anaconda_token: Add Anaconda.org token to automatically see private packages#

When the channel alias is Anaconda.org or an Anaconda Server GUI, you can set the system configuration so that users automatically see private packages. Anaconda.org was formerly known as binstar.org. This uses the Anaconda command-line client, which you can install with conda install anaconda-client, to automatically add the token to the channel URLs.

The default is True.

Example:

add_anaconda_token: False

Note

Even when set to True, this setting is enabled only if the Anaconda command-line client is installed and you are logged in with the anaconda login command.

envs_dirs: Specify environment directories#

Specify directories in which environments are located. If this key is set, the root prefix envs_dir is not used unless explicitly included. This key also determines where the package caches are located.

For each envs here, envs/pkgs is used as the pkgs cache, except for the standard envs directory in the root directory, for which the normal root_dir/pkgs is used.

Example:

envs_dirs:
  - ~/my-envs
  - /opt/anaconda/envs

The CONDA_ENVS_PATH environment variable overwrites the envs_dirs setting:

  • For macOS and Linux: CONDA_ENVS_PATH=~/my-envs:/opt/anaconda/envs

  • For Windows: set CONDA_ENVS_PATH=C:\Users\joe\envs;C:\Anaconda\envs

pkgs_dirs: Specify package directories#

Specify directories in which packages are located. If this key is set, the root prefix pkgs_dirs is not used unless explicitly included.

If the pkgs_dirs key is not set, then envs/pkgs is used as the pkgs cache, except for the standard envs directory in the root directory, for which the normal root_dir/pkgs is used.

Example:

pkgs_dirs:
  - /opt/anaconda/pkgs

The CONDA_PKGS_DIRS environment variable overwrites the pkgs_dirs setting:

  • For macOS and Linux: CONDA_PKGS_DIRS=/opt/anaconda/pkgs

  • For Windows: set CONDA_PKGS_DIRS=C:\Anaconda\pkgs

use_only_tar_bz2: Force conda to download only .tar.bz2 packages#

Conda 4.7 introduced a new .conda package file format. .conda is a more compact and faster alternative to .tar.bz2 packages. It's thus the preferred file format to use where available.

Nevertheless, it's possible to force conda to only download .tar.bz2 packages by setting the use_only_tar_bz2 boolean to True.

The default is False.

Example:

use_only_tar_bz2: True

Note

This is forced to True if conda-build is installed and older than 3.18.3, because older versions of conda break when conda feeds it the new file format.

Conda-build configuration#

root-dir: Specify conda-build output root directory#

Build output root directory. You can also set this with the CONDA_BLD_PATH environment variable. The default is <CONDA_PREFIX>/conda-bld/. If you do not have write permissions to <CONDA_PREFIX>/conda-bld/, the default is ~/conda-bld/.

Example:

conda-build:
    root-dir: ~/conda-builds

output_folder: Specify conda-build build folder (conda-build 3.16.3+)#

Folder to dump output package to. Packages are moved here if build or test succeeds. If unset, the output folder corresponds to the same directory as root-dir: the root build directory. .. code-block:: yaml

conda-build:

output_folder: conda-bld

pkg_version: Specify conda-build package version#

Conda package version to create. Use 2 for .conda packages. If not set, conda-build defaults to .tar.bz2.

conda-build:
   pkg_format: 2

anaconda_upload: Automatically upload conda-build packages to Anaconda.org#

Automatically upload packages built with conda-build to Anaconda.org. The default is False.

Example:

anaconda_upload: True

anaconda_token: Token to be used for Anaconda.org uploads (conda-build 3.0+)#

Tokens are a means of authenticating with Anaconda.org without logging in. You can pass your token to conda-build with this .condarc setting, or with a CLI argument. This is unset by default. Setting it implicitly enables anaconda_upload.

conda-build:
    anaconda_token: gobbledygook

quiet: Limit build output verbosity (conda-build 3.0+)#

Conda-build's output verbosity can be reduced with the quiet setting. For more verbosity, use the CLI flag --debug.

conda-build:
    quiet: true

filename_hashing: Disable filename hashing (conda-build 3.0+)#

Conda-build 3 adds hashes to filenames to allow greater customization of dependency versions. If you find this disruptive, you can disable the hashing with the following config entry:

conda-build:
    filename_hashing: false

Warning

Conda-build does not check when clobbering packages. If you utilize conda-build 3's build matrices with a build configuration that is not reflected in the build string, packages will be missing due to clobbering.

no_verify: Disable recipe and package verification (conda-build 3.0+)#

By default, conda-build uses conda-verify to ensure that your recipe and package meet some minimum sanity checks. You can disable these:

conda-build:
    no_verify: true

set_build_id: Disable per-build folder creation (conda-build 3.0+)#

By default, conda-build creates a new folder for each build, named for the package name plus a timestamp. This allows you to do multiple builds at once. If you have issues with long paths, you may need to disable this behavior. You should first try to change the build output root directory with the root-dir setting described above, but fall back to this as necessary:

conda-build:
    set_build_id: false

skip_existing: Skip building packages that already exist (conda-build 3.0+)#

By default, conda-build builds all recipes that you specify. You can instead skip recipes that are already built. A recipe is skipped if and only if all of its outputs are available on your currently configured channels.

conda-build:
    skip_existing: true

include_recipe: Omit recipe from package (conda-build 3.0+)#

By default, conda-build includes the recipe that was used to build the package. If this contains sensitive or proprietary information, you can omit the recipe.

conda-build:
    include_recipe: false

Note

If you do not include the recipe, you cannot use conda-build to test the package after the build completes. This means that you cannot split your build and test steps across two distinct CLI commands (conda build --notest recipe and conda build -t recipe). If you need to omit the recipe and split your steps, your only option is to remove the recipe files from the tarball artifacts after your test step. Conda-build does not provide tools for doing that.

activate: Disable activation of environments during build/test (conda-build 3.0+)#

By default, conda-build activates the build and test environments prior to executing the build or test scripts. This adds necessary PATH entries, and also runs any activate.d scripts you may have. If you disable activation, the PATH will still be modified, but the activate.d scripts will not run. This is not recommended, but some people prefer this.

conda-build:
    activate: false

long_test_prefix: Disable long prefix during test (conda-build 3.16.3+)#

By default, conda-build uses a long prefix for the test prefix. If you have recipes that fail in long prefixes but would still like to test them in short prefixes, you can disable the long test prefix. This is not recommended.

conda-build:
    long_test_prefix: false

The default is true.

pypirc: PyPI upload settings (conda-build 3.0+)#

Unset by default. If you have wheel outputs in your recipe, conda-build will try to upload them to the PyPI repository specified by the pypi_repository setting using credentials from this file path.

conda-build:
    pypirc: ~/.pypirc

pypi_repository: PyPI repository to upload to (conda-build 3.0+)#

Unset by default. If you have wheel outputs in your recipe, conda-build will try to upload them to this PyPI repository using credentials from the file specified by the pypirc setting.

conda-build:
    pypi_repository: pypi

Expansion of environment variables#

Conda expands environment variables in a subset of configuration settings. These are:

  • channel

  • channel_alias

  • channels

  • client_cert_key

  • client_cert

  • custom_channels

  • custom_multichannels

  • default_channels

  • envs_dirs

  • envs_path

  • migrated_custom_channels

  • pkgs_dirs

  • proxy_servers

  • verify_ssl

  • allowlist_channels

This allows you to store the credentials of a private repository in an environment variable, like so:

channels:
  - https://${USERNAME}:${PASSWORD}@my.private.conda.channel

Configuring number of threads#

You can use your .condarc file or environment variables to add configuration to control the number of threads. You may want to do this to tweak conda to better utilize your system. If you have a very fast SSD, you might increase the number of threads to shorten the time it takes for conda to create environments and install/remove packages.

repodata_threads#

  • Default number of threads: None

  • Threads used when downloading, parsing, and creating repodata structures from repodata.json files. Multiple downloads from different channels may occur simultaneously. This speeds up the time it takes to start solving.

verify_threads#

  • Default number of threads: 1

  • Threads used when verifying the integrity of packages and files to be installed in your environment. Defaults to 1, as using multiple threads here can run into problems with slower hard drives.

execute_threads#

  • Default number of threads: 1

  • Threads used to unlink, remove, link, or copy files into your environment. Defaults to 1, as using multiple threads here can run into problems with slower hard drives.

default_threads#

  • Default number of threads: None

  • When set, this value is used for all of the above thread settings. With its default setting (None), it does not affect the other settings.

Setting any of the above can be done in .condarc or with conda config:

At your terminal:

conda config --set repodata_threads 2

In .condarc:

verify_threads: 4