From ea751d5530a3dcb104773c4099009190cdf9dae7 Mon Sep 17 00:00:00 2001 From: felixfontein Date: Wed, 18 Jun 2025 19:53:53 +0000 Subject: [PATCH] deploy: c1df4bc8da24f9e020f2439a649e96ba6de1052a --- branch/main/docsite/scenario_guide.html | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/branch/main/docsite/scenario_guide.html b/branch/main/docsite/scenario_guide.html index f511b322..960a5895 100644 --- a/branch/main/docsite/scenario_guide.html +++ b/branch/main/docsite/scenario_guide.html @@ -197,11 +197,11 @@

Most of the modules and plugins in community.docker require the Docker SDK for Python. The SDK needs to be installed on the machines where the modules and plugins are executed, and for the Python version(s) with which the modules and plugins are executed. You can use the community.general.python_requirements_info module to make sure that the Docker SDK for Python is installed on the correct machine and for the Python version used by Ansible.

Note that plugins (inventory plugins and connection plugins) are always executed in the context of Ansible itself. If you use a plugin that requires the Docker SDK for Python, you need to install it on the machine running ansible or ansible-playbook and for the same Python interpreter used by Ansible. To see which Python is used, run ansible --version.

You can install the Docker SDK for Python for Python 3.6 or later as follows:

-
$ pip install docker
+
$ pip install docker
 

 

 
For Python 2.7, you need to use a version between 2.0.0 and 4.4.4 since the Python package for Docker removed support for Python 2.7 on 5.0.0. You can install the specific version of the Docker SDK for Python as follows:

-
$ pip install 'docker==4.4.4'
+
$ pip install 'docker==4.4.4'
 

 

 
Note that the Docker SDK for Python was called docker-py on PyPi before version 2.0.0. Please avoid installing this really old version, and make sure to not install both docker and docker-py. Installing both will result in a broken installation. If this happens, Ansible will detect it and inform you about it. If that happens, you must uninstall both and reinstall the correct version. If in doubt, always install docker and never docker-py.