The Cloud and local shells¶
To interact with the environment in your project, for example to run commands using its codebase, you can make use of a shell.
Once inside the shell, you can inspect the environment, run commands (such as
migrate). It’s particularly useful to be able to drop into a:
- Python console:
- Django shell:
python manage.py shell
- database shell:
python manage.py dbshell
The Cloud and local shells described here are not to be confused with the Divio Shell.
The Divio Shell is a convenient environment on your own computer, configured for interaction with your Divio account and projects, and the local development environment.
The Cloud and local shells will provide you with terminal access inside a running project.
Using a Cloud shell¶
The Cloud shell is only available for Managed Cloud projects.
Your Cloud server must be deployed in order to reach it via SSH. You can copy SSH connection details from the appropriate server pane in the Control Panel:
The command can be pasted into a terminal session on your own machine. An instance of your web
application will be spun up in a new container, and after a moment you’ll be logged in to it as
SSH sessions are limited to 30 minutes, regardless of any activity.
The Cloud shell instance¶
The container you’re connected to is a brand new instance. It will not be one actually serving your site on the web, but a new one that uses the same configuration, database and so on.
Files - any files you create or change on this instance will not affect those on any other containers.
Processes - each session is isolated from any other extant processes (web processes, workers, other shell sessions).
Caches - if your site’s cache relies on the database (the default in Divio projects) then
your container will be able to make use of it, clear it and so on. However, if you’re using for
locmem cache, it will not be available to your container.
The local shell¶
Open the shell¶
The Divio app provides a convenient short-cut in its toolbar:
Alternatively, you can use the command-line. In the project directory, run:
docker exec web bash # if the web container is already running
docker-compose run --rm web bash # if you need to start the container too