Page History

If you have opted for a custom cloud bundle variant, your site is hosted in Magnolia Cloud and you, or your cloud partner, manage:

...

must create and deploy the custom bundle

...

.

...

To successfully deploy custom bundles to Magnolia Cloud, certain best practices must be followed when creating your project and when deploying it the first time or continuously thereafter. 

Continuous delivery is a way to ensure that new configuration, templates and code can be rapidly and safely deployed to production. You deliver every change first to a UAT environment and verify that business applications and services function as expected through testing. Since every change is delivered to the UAT environment using automation in the Magnolia Cloud Cockpit, you have confidence that the application can be deployed to the Live environment exactly the same way, with a push of a button when the business is ready. In the UAT environment, business users can validate your changes with real content cloned from the Live environment. 
Adapted from Caum, Carl: Continuous Delivery Vs. Continuous Deployment: What's the Diff?

• First upgrade from standard bundle to a custom bundle is effort intensive on mgnl side due to dependency/compatibility issues. Only if data must be kept. Simpler alternative is to remove and replce when data doesn't need to be kept.
• Change in bundle name triggers same effort. +++Warning to not change name unless absolutely necessary.
• Partner bundle upgrade to newer partner bundle is fine.

Creating your custom bundle

...

• Basic checks for required configuration in the web.xml and magnolia.properties files.
• Inspection of required and blacklisted modules/jars.

Deployment of the

...

WAR file

The plugin uploads the built artifact to the entry point for Magnolia Cloud's deployment pipelines. It also collects meta-information about the artifact for debugging purposes.

Magnolia Cloud Simulator

For testing the We recommend you test your custom bundle in a simulated cloud environment before deploying it.

To do so, we offer a docker-compose based setup: Magnolia Cloud Simulator.