Exercise - Deploy a Spring Boot and Redis application to Azure App Service

  • 5 minutes

Now that you have everything running locally, let's deploy the application to the cloud and test its behavior in production.

Create an Azure App Service instance

Let's create an Azure App Service instance to host our application. The following code creates a basic App Service plan that uses Linux and then adds an instance of Java on App Service in that plan:

az appservice plan create \
    --resource-group $AZ_RESOURCE_GROUP \
    --name $AZ_REDIS_NAME-webapp-plan \
    --is-linux \
    --sku B1

az webapp create \
    --resource-group $AZ_RESOURCE_GROUP \
    --name $AZ_REDIS_NAME-webapp \
    --runtime "JAVA|8-jre8" \
    --plan $AZ_REDIS_NAME-webapp-plan

Note

If you'd rather use Java 11 instead of Java 8, use "JAVA|11-java11" as the web application runtime.

Note

We're reusing the $AZ_REDIS_NAME variable to name the App Service plan and the App Service instance, but you can use your own names here. "AZ" and the App Service instance name will be part of the URL, which needs to be unique across Azure.

You can go to the next section while this command finishes.

Deploy the Spring Boot application to Azure App Service

In your project's pom.xml file, add a new plug-in in the <plugins> section:

<plugin>
    <groupId>com.microsoft.azure</groupId>
    <artifactId>azure-webapp-maven-plugin</artifactId>
    <version>1.12.0</version>
    <configuration>
        <schemaVersion>V2</schemaVersion>
        <resourceGroup>${env.AZ_RESOURCE_GROUP}</resourceGroup>
        <appName>${env.AZ_REDIS_NAME}-webapp</appName>
        <deployment>
            <resources>
                <resource>
                    <directory>${project.basedir}/target</directory>
                    <includes>
                        <include>*.jar</include>
                    </includes>
                </resource>
            </resources>
        </deployment>
    </configuration>
</plugin>

Note

In the preceding configuration, you'll reuse the environment variables that you configured earlier. For this configuration to work, your $AZ_RESOURCE_GROUP and $AZ_REDIS_NAME variables need to be correctly set up.

You can now package your application and deploy it to Azure:

./mvnw package azure-webapp:deploy

That command will end up with a line like this one:

Successfully deployed the artifact to https://xxxx.azurewebsites.net

This is the URL of your running application, which we'll use in the next section.

Test the application on Azure

By using the URL from the previous section, you can now use the cURL commands that we were using on our local server, but on the cloud:

To retrieve data:

curl https://$AZ_REDIS_NAME-webapp.azurewebsites.net

To add an item:

curl -d '{"description":"a description", "details":"some details"}' -H "Content-Type: application/json" -X POST https://$AZ_REDIS_NAME-webapp.azurewebsites.net

And to increment your session:

curl -b cookie.txt -c cookie.txt https://$AZ_REDIS_NAME-webapp.azurewebsites.net/session

Note

Your local server and your cloud service use the same Redis instance. So they are, in fact, clustered! Any data that you write in one instance will be available in the other instance.

That's also how the cloud service will be able to scale out: by adding more nodes that will all use the same Redis instance to store both business data and session data.