Skip to content

Latest commit

 

History

History
1471 lines (1243 loc) · 57.5 KB

README.adoc

File metadata and controls

1471 lines (1243 loc) · 57.5 KB
Raw

servlet-async: How to Write an Asynchronous Servlet

The servlet-async quickstart demonstrates how to use asynchronous servlets to detach long-running tasks and free up the request processing thread.

What is it?

The servlet-async quickstart is a sample project showing the use of asynchronous servlets in WildFly Application Server.

It shows how to detach the execution of a long-running task from the request processing thread, so the thread is free to serve other client requests. The long-running tasks are executed using a dedicated thread pool and create the client response asynchronously.

A long-running task in this context does not refer to a computation intensive task executed on the same machine but could for example be contacting a third-party service that has limited resources or only allows for a limited number of concurrent connections. Moving the calls to this service into a separate and smaller sized thread pool ensures that less threads will be busy interacting with the long-running service and that more requests can be served that do not depend on this service.

System Requirements

The application this project produces is designed to be run on WildFly Application Server 34 or later.

All you need to build this project is Java SE 17.0 or later, and Maven 3.6.0 or later. See Configure Maven to Build and Deploy the Quickstarts to make sure you are configured correctly for testing the quickstarts.

Use of the WILDFLY_HOME and QUICKSTART_HOME Variables

In the following instructions, replace WILDFLY_HOME with the actual path to your WildFly installation. The installation path is described in detail here: Use of WILDFLY_HOME and JBOSS_HOME Variables.

When you see the replaceable variable QUICKSTART_HOME, replace it with the path to the root directory of all of the quickstarts.

Building and running the quickstart application with a WildFly server distribution

Start the WildFly Standalone Server

  1. Open a terminal and navigate to the root of the WildFly directory.

  2. Start the WildFly server with the default profile by typing the following command.

    $ WILDFLY_HOME/bin/standalone.sh
    Note
    		 For Windows, use the WILDFLY_HOME\bin\standalone.bat script.

Build and Deploy the Quickstart

  1. Make sure WildFly server is started.

  2. Open a terminal and navigate to the root directory of this quickstart.

  3. Type the following command to build the quickstart.

    $ mvn clean package

  4. Type the following command to deploy the quickstart.

    $ mvn wildfly:deploy

This deploys the servlet-async/target/servlet-async.war to the running instance of the server.

You should see a message in the server log indicating that the archive deployed successfully.

Access the Application

The application will be running at the following URL http://localhost:8080/servlet-async/.

Run the Integration Tests

This quickstart includes integration tests, which are located under the src/test/ directory. The integration tests verify that the quickstart runs correctly when deployed on the server.

Follow these steps to run the integration tests.

  1. Make sure WildFly server is started.

  2. Make sure the quickstart is deployed.

  3. Type the following command to run the verify goal with the integration-testing profile activated.

    $ mvn verify -Pintegration-testing

Undeploy the Quickstart

When you are finished testing the quickstart, follow these steps to undeploy the archive.

  1. Make sure WildFly server is started.

  2. Open a terminal and navigate to the root directory of this quickstart.

  3. Type this command to undeploy the archive:

    $ mvn wildfly:undeploy

Building and running the quickstart application with provisioned WildFly server

Instead of using a standard WildFly server distribution, you can alternatively provision a WildFly server to deploy and run the quickstart. The functionality is provided by the WildFly Maven Plugin, and you may find its configuration in the quickstart pom.xml:

        <profile>
            <id>provisioned-server</id>
            <activation>
                <activeByDefault>true</activeByDefault>
            </activation>
            <build>
                <plugins>
                    <plugin>
                        <groupId>org.wildfly.plugins</groupId>
                        <artifactId>wildfly-maven-plugin</artifactId>
                        <configuration>
                            <discover-provisioning-info>
                                <version>${version.server}</version>
                            </discover-provisioning-info>
                            <add-ons>...</add-ons>
                        </configuration>
                        <executions>
                            <execution>
                                <goals>
                                    <goal>package</goal>
                                </goals>
                            </execution>
                        </executions>
                    </plugin>
                    ...
                </plugins>
            </build>
        </profile>

When built, the provisioned WildFly server can be found in the target/server directory, and its usage is similar to a standard server distribution, with the simplification that there is never the need to specify the server configuration to be started.

Follow these steps to run the quickstart using the provisioned server.

Procedure

  1. Make sure the server is provisioned.

    $ mvn clean package

  2. Start the WildFly provisioned server, using the WildFly Maven Plugin start goal.

    $ mvn wildfly:start

  3. Type the following command to run the integration tests.

    $ mvn verify -Pintegration-testing

  4. Shut down the WildFly provisioned server.

    $ mvn wildfly:shutdown

Building and running the quickstart application with OpenShift

Build the WildFly Source-to-Image (S2I) Quickstart to OpenShift with Helm Charts

On OpenShift, the S2I build with Apache Maven uses an openshift Maven profile to provision a WildFly server, deploy and run the quickstart in OpenShift environment.

The server provisioning functionality is provided by the WildFly Maven Plugin, and you may find its configuration in the quickstart pom.xml:

        <profile>
            <id>openshift</id>
            <build>
                <plugins>
                    <plugin>
                        <groupId>org.wildfly.plugins</groupId>
                        <artifactId>wildfly-maven-plugin</artifactId>
                        <configuration>
                            <discover-provisioning-info>
                                <version>${version.server}</version>
                                <context>cloud</context>
                            </discover-provisioning-info>
                            <add-ons>...</add-ons>
                        </configuration>
                        <executions>
                            <execution>
                                <goals>
                                    <goal>package</goal>
                                </goals>
                            </execution>
                        </executions>
                    </plugin>
                    ...
                </plugins>
            </build>
        </profile>

You may note that unlike the provisioned-server profile it uses the cloud context which enables a configuration tuned for OpenShift environment.

The plugin uses WildFly Glow to discover the feature packs and layers required to run the application, and provisions a server containing those layers.

If you get an error or the server is missing some functionality which cannot be auto-discovered, you can download the WildFly Glow CLI and run the following command to see more information about what add-ons are available:

wildfly-glow show-add-ons

Getting Started with WildFly for OpenShift and Helm Charts

This section contains the basic instructions to build and deploy this quickstart to WildFly for OpenShift or WildFly for OpenShift Online using Helm Charts.

Prerequisites

  • You must be logged in OpenShift and have an oc client to connect to OpenShift

  • Helm must be installed to deploy the backend on OpenShift.

Once you have installed Helm, you need to add the repository that provides Helm Charts for WildFly.

$ helm repo add wildfly https://docs.wildfly.org/wildfly-charts/
"wildfly" has been added to your repositories
$ helm search repo wildfly
NAME                    CHART VERSION   APP VERSION     DESCRIPTION
wildfly/wildfly         ...             ...            Build and Deploy WildFly applications on OpenShift
wildfly/wildfly-common  ...             ...            A library chart for WildFly-based applications

Deploy the WildFly Source-to-Image (S2I) Quickstart to OpenShift with Helm Charts

Log in to your OpenShift instance using the oc login command. The backend will be built and deployed on OpenShift with a Helm Chart for WildFly.

Navigate to the root directory of this quickstart and run the following command:

$ helm install servlet-async -f charts/helm.yaml wildfly/wildfly --wait --timeout=10m0s 
NAME: servlet-async
...
STATUS: deployed
REVISION: 1

This command will return once the application has successfully deployed. In case of a timeout, you can check the status of the application with the following command in another terminal:

oc get deployment servlet-async

The Helm Chart for this quickstart contains all the information to build an image from the source code using S2I on Java 17:

build:
  uri: https://github.com/wildfly/quickstart.git
  ref: main
  contextDir: servlet-async
deploy:
  replicas: 2

This will create a new deployment on OpenShift and deploy the application.

If you want to see all the configuration elements to customize your deployment you can use the following command:

$ helm show readme wildfly/wildfly

Get the URL of the route to the deployment.

$ oc get route servlet-async -o jsonpath="{.spec.host}"

Access the application in your web browser using the displayed URL.

Run the Integration Tests with OpenShift

The integration tests included with this quickstart, which verify that the quickstart runs correctly, may also be run with the quickstart running on OpenShift.

Note

The integration tests expect a deployed application, so make sure you have deployed the quickstart on OpenShift before you begin.

Run the integration tests using the following command to run the verify goal with the integration-testing profile activated and the proper URL:

$ mvn verify -Pintegration-testing -Dserver.host=https://$(oc get route servlet-async --template='{{ .spec.host }}')
Note

The tests are using SSL to connect to the quickstart running on OpenShift. So you need the certificates to be trusted by the machine the tests are run from.

Undeploy the WildFly Source-to-Image (S2I) Quickstart from OpenShift with Helm Charts

$ helm uninstall servlet-async

Building and running the quickstart application with Kubernetes

Build the WildFly Quickstart to Kubernetes with Helm Charts

For Kubernetes, the build with Apache Maven uses an openshift Maven profile to provision a WildFly server, suitable for running on Kubernetes.

The server provisioning functionality is provided by the WildFly Maven Plugin, and you may find its configuration in the quickstart pom.xml:

        <profile>
            <id>openshift</id>
            <build>
                <plugins>
                    <plugin>
                        <groupId>org.wildfly.plugins</groupId>
                        <artifactId>wildfly-maven-plugin</artifactId>
                        <configuration>
                            <discover-provisioning-info>
                                <version>${version.server}</version>
                                <context>cloud</context>
                            </discover-provisioning-info>
                            <add-ons>...</add-ons>
                        </configuration>
                        <executions>
                            <execution>
                                <goals>
                                    <goal>package</goal>
                                </goals>
                            </execution>
                        </executions>
                    </plugin>
                    ...
                </plugins>
            </build>
        </profile>

You may note that unlike the provisioned-server profile it uses the cloud context which enables a configuration tuned for Kubernetes environment.

The plugin uses WildFly Glow to discover the feature packs and layers required to run the application, and provisions a server containing those layers.

If you get an error or the server is missing some functionality which cannot be auto-discovered, you can download the WildFly Glow CLI and run the following command to see more information about what add-ons are available:

wildfly-glow show-add-ons

Getting Started with Kubernetes and Helm Charts

This section contains the basic instructions to build and deploy this quickstart to Kubernetes using Helm Charts.

Install Kubernetes

In this example we are using Minikube as our Kubernetes provider. See the Minikube Getting Started guide for how to install it. After installing it, we start it with 4GB of memory.

minikube start --memory='4gb'

The above command should work if you have Docker installed on your machine. If, you are using Podman instead of Docker, you will also need to pass in --driver=podman, as covered in the Minikube documentation.

Once Minikube has started, we need to enable its registry since that is where we will push the image needed to deploy the quickstart, and where we will tell the Helm charts to download it from.

minikube addons enable registry

In order to be able to push images to the registry we need to make it accessible from outside Kubernetes. How we do this depends on your operating system. All the below examples will expose it at localhost:5000

# On Mac:
docker run --rm -it --network=host alpine ash -c "apk add socat && socat TCP-LISTEN:5000,reuseaddr,fork TCP:$(minikube ip):5000"

# On Linux:
kubectl port-forward --namespace kube-system service/registry 5000:80 &

# On Windows:
kubectl port-forward --namespace kube-system service/registry 5000:80
docker run --rm -it --network=host alpine ash -c "apk add socat && socat TCP-LISTEN:5000,reuseaddr,fork TCP:host.docker.internal:5000"

Prerequisites

  • Helm must be installed to deploy the backend on Kubernetes.

Once you have installed Helm, you need to add the repository that provides Helm Charts for WildFly.

$ helm repo add wildfly https://docs.wildfly.org/wildfly-charts/
"wildfly" has been added to your repositories
$ helm search repo wildfly
NAME                    CHART VERSION   APP VERSION     DESCRIPTION
wildfly/wildfly         ...             ...            Build and Deploy WildFly applications on OpenShift
wildfly/wildfly-common  ...             ...            A library chart for WildFly-based applications

Deploy the WildFly Source-to-Image (S2I) Quickstart to Kubernetes with Helm Charts

The backend will be built and deployed on Kubernetes with a Helm Chart for WildFly.

Navigate to the root directory of this quickstart and run the following commands:

mvn -Popenshift package wildfly:image

This will use the openshift Maven profile we saw earlier to build the application, and create a Docker image containing the WildFly server with the application deployed. The name of the image will be servlet-async.

Next we need to tag the image and make it available to Kubernetes. You can push it to a registry like quay.io. In this case we tag as localhost:5000/servlet-async:latest and push it to the internal registry in our Kubernetes instance:

# Tag the image
docker tag servlet-async localhost:5000/servlet-async:latest
# Push the image to the registry
docker push localhost:5000/servlet-async:latest

In the below call to helm install which deploys our application to Kubernetes, we are passing in some extra arguments to tweak the Helm build:

  • --set build.enabled=false - This turns off the s2i build for the Helm chart since Kubernetes, unlike OpenShift, does not have s2i. Instead, we are providing the image to use.

  • --set deploy.route.enabled=false - This disables route creation normally performed by the Helm chart. On Kubernetes we will use port-forwards instead to access our application, since routes are an OpenShift specific concept and thus not available on Kubernetes.

  • --set image.name="localhost:5000/servlet-async" - This tells the Helm chart to use the image we built, tagged and pushed to Kubernetes' internal registry above.

$ helm install servlet-async -f charts/helm.yaml wildfly/wildfly --wait --timeout=10m0s --set build.enabled=false --set deploy.route.enabled=false --set image.name="localhost:5000/servlet-async"
NAME: servlet-async
...
STATUS: deployed
REVISION: 1

This command will return once the application has successfully deployed. In case of a timeout, you can check the status of the application with the following command in another terminal:

kubectl get deployment servlet-async

The Helm Chart for this quickstart contains all the information to build an image from the source code using S2I on Java 17:

build:
  uri: https://github.com/wildfly/quickstart.git
  ref: main
  contextDir: servlet-async
deploy:
  replicas: 2

This will create a new deployment on Kubernetes and deploy the application.

If you want to see all the configuration elements to customize your deployment you can use the following command:

$ helm show readme wildfly/wildfly

To be able to connect to our application running in Kubernetes from outside, we need to set up a port-forward to the servlet-async service created for us by the Helm chart.

This service will run on port 8080, and we set up the port forward to also run on port 8080:

kubectl port-forward service/servlet-async 8080:8080

The server can now be accessed via http://localhost:8080 from outside Kubernetes. Note that the command to create the port-forward will not return, so it is easiest to run this in a separate terminal.

Run the Integration Tests with Kubernetes

The integration tests included with this quickstart, which verify that the quickstart runs correctly, may also be run with the quickstart running on Kubernetes.

Note

The integration tests expect a deployed application, so make sure you have deployed the quickstart on Kubernetes before you begin.

Run the integration tests using the following command to run the verify goal with the integration-testing profile activated and the proper URL:

$ mvn verify -Pintegration-testing -Dserver.host=http://localhost:8080

Undeploy the WildFly Source-to-Image (S2I) Quickstart from Kubernetes with Helm Charts

$ helm uninstall servlet-async

To stop the port forward you created earlier use:

$ kubectl port-forward service/servlet-async 8080:8080