Azure Service Fabric Java Client APIs

Service Fabric client APIs allows deploying and managing microservices based applications and containers in a Service Fabric cluster on Azure, on-premises, on local development machine or in other cloud. This article describes how to generate and use Service Fabric Java client APIs on top of the Service Fabric client REST APIs

Generate the client code using AutoRest

AutoRest is a tool that generates client libraries for accessing RESTful web services. Input to AutoRest is a specification that describes the REST API using the OpenAPI Specification format. Service Fabric client REST APIs follow this specification .

Follow the steps mentioned below to generate Service Fabric Java client code using the AutoRest tool.

1. Install nodejs and NPM on your machine

   If you are using Linux then:

   sudo apt-get install npm
   sudo apt install nodejs
   

   If you are using Mac OS X then:

   brew install node
   

2. Install AutoRest using NPM.

   npm install -g autorest
   

3. Fork and clone azure-rest-api-specs repository in your local machine and go to the cloned location from the terminal of your machine.

4. Go to the location mentioned below in your cloned repo.

   cd specification\servicefabric\data-plane\Microsoft.ServiceFabric\stable\6.0
   

   Note

   If your cluster version is not 6.0.* then go to the appropriate directory in the stable folder.

5. Run the following autorest command to generate the Java client code.

   autorest --input-file= servicefabric.json --java --output-folder=[output-folder-name] --namespace=[namespace-of-generated-client]
   

   Below is an example demonstrating the usage of autorest.

   autorest --input-file=servicefabric.json --java --output-folder=java-rest-api-code --namespace=servicefabricrest
   

   The following command takes servicefabric.json specification file as input and generates Java client code in java-rest-api-code folder and encloses the code in servicefabricrest namespace. After this step you would find two folders models, implementation and two files ServiceFabricClientAPIs.java and package-info.java generated in the java-rest-api-code folder.

Include and use the generated client in your project

1. Add the generated code appropriately into your project. We recommend that you create a library using the generated code and include this library in your project.

2. If you are creating a library then include the following dependency in your library's project. If you are following a different approach then include the dependency appropriately.

   GroupId:  com.microsoft.rest
   Artifactid: client-runtime
   Version: 1.2.1
   

   For example, if you are using Maven build system include the following in your pom.xml file:

       <dependency>
         <groupId>com.microsoft.rest</groupId>
         <artifactId>client-runtime</artifactId>
         <version>1.2.1</version>
       </dependency>
   

3. Create a RestClient using the following code:

       RestClient simpleClient = new RestClient.Builder()
           .withBaseUrl("http://<cluster-ip or name:port>")
           .withResponseBuilderFactory(new ServiceResponseBuilder.Factory())
           .withSerializerAdapter(new JacksonAdapter())
           .build();
       ServiceFabricClientAPIs client = new ServiceFabricClientAPIsImpl(simpleClient);
   

4. Use the client object and make the appropriate calls as required. Here are some examples which demonstrate the usage of client object. We assume that the application package is built and uploaded into image store before using the below API's.

   • Provision an application

         ApplicationTypeImageStorePath imageStorePath = new ApplicationTypeImageStorePath();
         imageStorePath.withApplicationTypeBuildPath("<application-path-in-image-store>");
         client.provisionApplicationType(imageStorePath);
     

   • Create an application

         ApplicationDescription applicationDescription = new ApplicationDescription();
         applicationDescription.withName("<application-uri>");
         applicationDescription.withTypeName("<application-type>");
         applicationDescription.withTypeVersion("<application-version>");
         client.createApplication(applicationDescription);
     

Understanding the generated code

For every API you will find four overloads of implementation. If there are optional parameters then you would find four more variations including those optional parameters. For example consider the API removeReplica.

1. public void removeReplica(String nodeName, UUID partitionId, String replicaId, Boolean forceRemove, Long timeout)
   • This is the synchronous variant of the removeReplica API call
2. public ServiceFuture<Void> removeReplicaAsync(String nodeName, UUID partitionId, String replicaId, Boolean forceRemove, Long timeout, final ServiceCallback<Void> serviceCallback)
   • This variant of API call can be used if you want to use future based asynchronous programming and use callbacks
3. public Observable<Void> removeReplicaAsync(String nodeName, UUID partitionId, String replicaId)
   • This variant of API call can be used if you want to use reactive asynchronous programming
4. public Observable<ServiceResponse<Void>> removeReplicaWithServiceResponseAsync(String nodeName, UUID partitionId, String replicaId)
   • This variant of API call can be used if you want to use reactive asynchronous programming and deal with RAW rest response

Next steps

• Learn about Service Fabric REST APIs