Skip to content

Latest commit

 

History

History
304 lines (229 loc) · 10.9 KB

README.adoc

File metadata and controls

304 lines (229 loc) · 10.9 KB
Raw

todo-backend: quickstart for backend deployment on OpenShift

The todo-backend quickstart demonstrates how to implement a backend that exposes a HTTP API with JAX-RS to manage a list of ToDo which are persisted in a database with JPA.

This quickstart shows how to setup a local deployment of this backend as well as a deployment on OpenShift to connect to a PostgreSQL database also hosted on OpenShift.

What is it?

The todo-back quickstart demonstrates how to implement a backend that exposes a HTTP API with JAX-RS to manage a list of ToDo which are persisted in a database with JPA.

  • The backend exposes a HTTP API to manage a list of todos that complies with the specs defined at todobackend.com.

  • It requires a connection to a PostgreSQL database to persist the todos.

  • It uses the Bootable Jar for local and cloud deployment

  • It is deployed on OpenShift using the Helm Chart for ${productName}.

Architecture

This backend is built and deployed as Bootable Jar that provisions the {productName} application server and all the feature packs it needs for its features. The layers are defined in the pom.xml file in the <configuration> section of the org.wildfly.plugins:wildfly-jar-maven-plugin plugin:

<layers>
  <layer>cloud-server</layer>
  <layer>postgresql-datasource</layer>
  <layer>ejb</layer>
</layers>

The cloud-server and ejb layers provides everything need to run the backend on OpenShift. This also includes access to Jakarta EE APIs such as CDI, JAX-RS, JPA, etc. These two layers comes from the {productName} feature pack provided by the Bootable Jar plugin:

<feature-pack>
  <location>wildfly@maven(org.jboss.universe:community-universe)#${version.server.bootable-jar}</location>
</feature-pack>

The postgresql-datasource layer provides a JDBC driver and DataSource to connect to a PostgreSQL database. It is not provided by the {productName} feature pack but by an extra feature pack:

 <feature-pack>
   <groupId>org.wildfly</groupId>
   <artifactId>wildfly-datasources-galleon-pack</artifactId>
   <version>${version.wildfly-datasources-galleon-pack}</version>
 </feature-pack>

The Git repository for this feature pack is hosted at https://github.com/wildfly-extras/wildfly-datasources-galleon-pack. It provides JDBC drivers and datasources for different databases but for this quickstart, we will only need the postgresql-datasource.

Connection to the PostgreSQL database

As mentioned, the JDBC drivers and datasource configuration that the backend uses to connect to the PostgreSQL database is provided by the org.wildfly:wildfly-datasources-galleon-pack feature pack. By default, it exposes a single datasource. In the backend, the name of this datasource is ToDos and is specified in the persistence.xml to configure JPA:

<persistence-unit name="primary">
  <jta-data-source>java:jboss/datasources/ToDos</jta-data-source>
</persistence-unit>

At runtime, we only need a few environment variables to establish the connection from {productName} to the external PostgreSQL database:

  • POSTGRESQL_DATABASE - the name of the database (that will be called todos)

  • POSTGRESQL_SERVICE_HOST - the host to connect to the database

  • POSTGRESQL_SERVICE_PORT - The port to connecto to the database

  • POSTGRESQL_USER & POSTGRESQL_PASSWORD - the credentials to connect to the database

  • POSTGRESQL_DATASOURCE - The name of the datasourcs (as mentioned above, it will be ToDos)

Filters for Cross-Origin Resource Sharing (CORS)

The Web frontend for this quickstart uses JavaScript calls to query the backend’s HTTP API. We must enable Cross-Origin Resource Sharing (CORS) filters in the undertow subsystem of {productName}.

As we use Bootable Jar to build the application, we provide a CLI script that contains all the commands to create and configure the CORS filters in Undertow. This script is located in the src/scripts/cors_filters.cli.

This script is executed at build time and will provide the following HTTP headers to enabled CORS:

  • Access-Control-Allow-Origin: *

  • Access-Control-Allow-Methods: GET, POST, OPTION, PUT, DELETE, PATCH

  • Access-Control-Allow-Headers: accept, authorization, content-type, x-requested-with

  • Access-Control-Allow-Credentials: true

  • Access-Control-Max-Age: 1

By default, the backend accepts requests from any origin (*`). This is only simplicity. It is possible to restrict the allowed origin using the environment variable CORS_ORIGIN at runtime.

Package the backend as a Bootable Jar

The backend is packaged as a Bootable Jar and configured to be deployable on OpenShift with the Maven Profile bootable-jar-openshift which is active by default:

$ mvn clean package

Run the backend locally

Local PostgreSQL database

Before running the backend locally, we need to have a local PostgreSQL database that we can connect to. We use the postgresql docker image to create one:

$ docker run --name todos-db \
  -e POSTGRES_USER=todos \
  -e POSTGRES_PASSWORD=mysecretpassword \
  -p 5432:5432 \
  postgres

This will create a database named todos-db that we can connect to on localhost:5432 with the credentials todos / mysecretpassword.

Run the backend

With the PostgreSQL database running, we can start the backend by passing the required environment variables to connect to the database:

$ POSTGRESQL_DATABASE=todos \
  POSTGRESQL_SERVICE_HOST=localhost \
  POSTGRESQL_SERVICE_PORT=5432 \
  POSTGRESQL_USER=todos \
  POSTGRESQL_PASSWORD=mysecretpassword \
  POSTGRESQL_DATASOURCE=ToDos \
  java -jar target/todo-backend-bootable.jar
...
14:41:58,111 INFO  [org.jboss.as.server] (Controller Boot Thread) WFLYSRV0010: Deployed "todo-backend.war" (runtime-name : "ROOT.war")
...

The backend is running, and we can use the HTTP API to manage a list of todos:

# get a list of todos
$ curl http://localhost:8080
[]

# create a todo with the title "This is my first todo item!"
$ curl -X POST -H "Content-Type: application/json"  -d '{"title": "This is my first todo item!"}' http://localhost:8080/
{"completed":false,"id":1,"order":0,"title":"This is my first todo item!","url":"https://localhost:8080/1"}%

# get a list of todos with the one that was just created
$ curl http://localhost:8080
[{"completed":false,"id":1,"order":0,"title":"This is my first todo item!","url":"https://localhost:8080/1"}

Run the Backend on OpenShift

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 {productName}:

$ 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         1.1.0           22.0            Build and Deploy WildFly applications on OpenShift

Deploy a PostgreSQL Database on OpenShift

$ oc new-app postgresql-ephemeral \
   -p DATABASE_SERVICE_NAME=todos-db \
   -p POSTGRESQL_DATABASE=todos

This will create a PostgreSQL database named todos on OpenShift that can be accessed on the port 5432 on the service todos-db. We don’t need to copy the credentials to connect to the database as we will retrieve them later using the todos-db secret that was created when the database is deployed.

Build and Deploy the Backend on OpenShift

The backend will be built and deployed on OpenShift with a Helm Chart for {productName}.

$ helm install todo-backend -f todo-backend.yaml wildfly/wildfly
NAME: todo-backend
...
STATUS: deployed
REVISION: 1

The Helm Chart for this quickstart contains all the information to build an image from the source code using Bootable Jar:

build:
  uri: https://github.com/wildfly/quickstart.git
  mode: bootable-jar

The Helm Chart also contains the environment variables required to connect to the PostgreSQL database. In local deployment the credentials were passed directly as the values of the environment variables. For OpenShift, we rely on secrets so that the credentials are never copied outside OpenShift:

deploy:
  env:
    - name: POSTGRESQL_PASSWORD
      valueFrom:
        secretKeyRef:
          key: database-password
          name: todos-db

When the application is deployed, the value for the POSTGRESQL_PASSWORD will be taken from the key database-password in the secrets todos-db.

Let’s wait for the application to be built and deployed:

$ oc get dc/todo-backend -w
NAME           REVISION   DESIRED   CURRENT   TRIGGERED BY
todo-backend   0          3         0         config,image(todo-backend:latest)
...
todo-backend   1          3         3         config,image(todo-backend:latest)

Use the todobackend Web application

Once the backend is deployed on OpenShift, it can be accessed from the route todo-backend. Let’s find the host that we can use to connect to this backend:

$ oc get route todo-backend -o jsonpath="{.spec.host}"
todo-backend-jmesnil1-dev.apps.sandbox.x8i5.p1.openshiftapps.com

This value will be different for every installation of the backend.

Warning

Make sure to prepend the host with https:// to be able to connect to the backend from the ToDo Backend Specs or Client

We can verify that this application is properly working as a ToDo Backend by running its specs on it.

Once all tests passed, we can use the todobackend client to have a Web application connected to the backend.

Clean up

Remove the Backend

The backend can be deleted from OpenShift by running the command:

$ helm delete todo-backend
release "todo-backend" uninstalled

Remove the Database

The PostresSQL database can be deleted from OpenShift by running the commands:

$ oc delete all -l template=postgresql-ephemeral-template
replicationcontroller "todos-db-1" deleted
service "todos-db" deleted
deploymentconfig.apps.openshift.io "todos-db" deleted
$ oc delete secret todos-db
secret "todos-db" deleted