CloudLaunch is a ReSTful, extensible Django app for discovering and launching applications on cloud, container, or local infrastructure. A live version is available at https://launch.usegalaxy.org/.
CloudLaunch can be extended with your own plug-ins, which can provide custom launch logic for arbitrary applications. Visit the live site to see currently available applications in the Catalog. CloudLaunch is also tightly integrated with CloudBridge, which makes CloudLaunch natively multi-cloud. If you would like to have an additional cloud provider added as an available option for a given appliance, please create an issue in this repo.
CloudLaunch has a web and commandline front-end. The Web UI is maintained in the CloudLaunch-UI repository. The commandline client is maintained in the cloudlaunch-cli repository.
The recommended way to install CloudLaunch is via the CloudLaunch Helm Chart: https://github.com/cloudve/cloudlaunch-helm
- Install the CloudLaunch Django server
$ pip install cloudlaunch-server
Once installed, you can run Django admin commands as follows:
$ cloudlaunch-server django
- Copy
cloudlaunchserver/settings_local.py.sample
tocloudlaunchserver/settings_local.py
and make any desired configuration changes. Make sure to change the value forFERNET_KEYS
variable because it is used to encrypt sensitive database fields. - Prepare the database with:
$ cloudlaunch-server django migrate
$ cloudlaunch-server django createsuperuser
- Start the development server and celery task queue (along with a Redis server as the message broker), each process in its own tab.
$ python manage.py runserver
$ redis-server & celery -A cloudlaunchserver worker -l info --beat
- Visit http://127.0.0.1:8000/cloudlaunch/admin/ to define your application and infrastructure properties.
- 6 . Install the UI for the server by following instructions from
- https://github.com/galaxyproject/cloudlaunch-ui.
CloudLaunch is based on Python 3.6 and although it may work on older Python versions, 3.6 is the only supported version. Use of Conda or virtualenv is also highly advised.
- Checkout CloudLaunch and create an isolated environment
$ conda create --name cl --yes python=3.6
$ conda activate cl
$ git clone https://github.com/galaxyproject/cloudlaunch.git
$ cd cloudlaunch
$ pip install -r requirements_dev.txt
$ cd django-cloudlaunch
- Copy
cloudlaunchserver/settings_local.py.sample
tocloudlaunchserver/settings_local.py
and make any desired configuration changes. - Run the migrations and create a superuser:
$ python manage.py migrate
$ python manage.py createsuperuser
- Start the web server and Celery in separate tabs
$ python manage.py runserver
$ redis-server & celery -A cloudlaunchserver worker -l info --beat
- Visit http://127.0.0.1:8000/cloudlaunch/admin/ to define appliances and add cloud providers.
- Visit http://127.0.0.1:8000/cloudlaunch/api/v1/ to explore the API.
- 7 . Install the UI for the server by following instructions from
- https://github.com/galaxyproject/cloudlaunch-ui.
Every PR should also bump the version or build number. Do this by running one of the following commands as part of the PR, which will create a commit:
- For updating a dev version:
bumpversion [major | minor | patch]
eg, with current version 4.0.0, runningbumpversion patch
will result in 4.0.1-dev0 - For updating a build version:
bumpversion build
will result in 4.0.1-dev1 - For production version:
bumpversion --tag release
will result in 4.0.1, with a git tag