Defining a Remote Server Connection via Swordfish Tooling

Swordfish provides local and remote server deployment of Swordfish features, that is, a Swordfish Service Provider created within the Ecipse IDE Workbench can be deployed onto a remote Swordfish Runtime Server (OSGi enabled). The client deployer functionality allows you to define the connection properties to a remote server via a the Eclipse IDE,and then add features to this newly defined remote server.

These are the main tasks to be completed to deploy features onto a remote server.


  1. Create plug-in projects
  2. Create feature plug-in development projects and asisgn plug-ins to them.
  3. Convert the feature plug-in development projects to faceted form and assign them to Swordfish Runtime
  4. Add a new Swordfish Server definition.
  5. Add the faceted feature projects to the the server and publish them.
Note: To avoid dependency problems, transfer all features that may contain dependent plug-ins in a single step.

The client side deployer can deploy any set of plug-ins that are represented as plug-in projects in the Eclipse IDE. Plug-ins that are to be deployed must be assigned to features. The assignment does not need to be exclusive. This means that some plug-ins can be contained within multiple features. Multiple features may be deployed in one deployment step. The order of deployment is important when there are dependencies between the plug-ins. During deployment the plug-ins of all features marked to be deployed are transferred to the server. Then the P2 planner is used to decide whether all mutual plug-in dependencies are fulfilled by the transferred plug-ins in relation to the plug-ins that already running on the server. If not, the deployment is rolled back and an error message referencing the first unfulfilled dependency is displayed.

Instructions for each of these tasks can be found in this help, see Swordfish Deployment Tasks.