mpbauer / spring-petclinic-rest-slim

Slimmed down REST version of the Spring Petclinic sample application

Spring PetClinic Sample Application (Slim)

This backend version of the Spring Petclinic application only provides a REST API. There is no UI. The spring-petclinic-angular project is a Angular front-end application which consumes the REST API.

A modified version (specific to this project) of the spring-petclinic-angular project can be found HERE

---

NOTE

This project is based upon code from the commit reference f97f1d44580d9950463043b451902f2e4955dea7 of the repository spring-petclinic/spring-petclinic-rest.

---

Understanding the Spring Petclinic application with a few diagrams

See the presentation of the Spring Petclinic Framework version

Petclinic ER Model

Running the petclinic application locally

With Maven

./mvnw spring-boot:run

Packaging and running the application with JVM

The application ca be packaged using:

./mnvw package

It produces the spring-petclinic-rest-slim-{{version}}.jar file in the /target directory. The jar is already a fat-jar and can be executed with the following command:

java -jar ./target/spring-petclinic-rest-slim-{{version}}.jar

With Docker

Before building the container image run:

./mvnw package

Then, build the image with:

docker build -f src/main/docker/Dockerfile.jvm -t mpbauer/spring-petclinic-rest-slim-jvm .

Then run the container using:

docker run -i --rm -p 8080:8080 mpbauer/spring-petclinic-rest-slim-jvm

You can then access petclinic here: http://localhost:8080/petclinic/

Creating a native executable

Spring Native provides support to compile Spring application to a native executable by utilizing GraalVM.

🚧 Warning: This section is under heavy construction

At the time of writing the spring-native project is still in alpha and the number of supported dependencies is still relatively small. A native configuration will be implemented once the spring-native project is more stable and supports a wider range of dependencies.

Open API REST Documentation

The following URLs can be used to access a documentation about the spring-petclinic-rest-slim application:

Swagger UI

http://localhost:8080/petclinic/swagger-ui

Open API Schema Document

http://localhost:8080/petclinic/v3/api-docs

Health Checks

The spring-boot-starter-actuator dependency provides a health check endpoint at /actuator/health

Example:

http://localhost:8080/petclinic/actuator/health

Metrics

The spring-boot-starter-actuator provides two metric endpoints:

• /actuator/metrics - Shows ‘metrics’ information for the current application.
• /actuator/prometheus - Exposes metrics in a format that can be scraped by a Prometheus server. Requires a dependency on micrometer-registry-prometheus

Metrics endpoint:

http://localhost:8080/petclinic/actuator/metrics

Prometheus endpoint:

http://localhost:8080/petclinic/actuator/prometheus

Database configuration

The database support for this version of the spring-petlinic-rest project was significantly reduced. As of now this project only supports PostgreSQL and H2.

In its default configuration a PostgreSQL database is required to run the application. For the execution of tests an embedded H2 is started.

For local development you may want to start a PostgreSQL database with Docker:

docker run --name petclinic -p 5432:5432 -e POSTGRES_PASSWORD=pass -d postgres

Security configuration

A Role Based Access Control is enabled by default when running the application with the default profile. When you start the application with the local profile through ./mvnw spring-boot:run -Dspring-boot.run.profiles=local authentication is disabled by default.

At the moment JWT based authentication is not implemented yet

Enable Authentication

In order to use the JWT based authentication functionality, you can turn it on by setting the following property in the application.properties file:

petclinic.security.enable=true

Authorization

This will secure all APIs and in order to access them, basic authentication is required. Apart from authentication, APIs also require authorization. This is done via roles that a user can have. The existing roles are listed below with the corresponding permissions:

Role	Controller
OWNER_ADMIN	OwnerController PetController PetTypeController (getAllPetTypes() & getPetType())
VET_ADMIN	PetTypeController SpecialityController VetController
ADMIN	UserController

Roles Based Access Control with predefined JSON Web Tokens

To make the use of this sample application as easy as possible a set of fixed JSON Web Tokens with different roles were generated and signed with a secret that is hardcoded in application.properties. You can copy the pre-created tokens and pass it via the Authorization header to the application.

Example:

curl --location --request GET 'http://localhost:8080/petclinic/api/owners' \
--header 'Content-Type: application/json' \
--header 'Accept: application/json' \
--header 'Authorization: Bearer eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJYMDAwMDQiLCJpc3MiOiJodHRwczovL3NwcmluZy1wZXRjbGluaWMuZ2l0aHViLmlvL2lzc3VlciIsImdyb3VwcyI6WyJST0xFX09XTkVSX0FETUlOIiwiUk9MRV9WRVRfQURNSU4iLCJST0xFX0FETUlOIl0sImV4cCI6NDc3MDU1NDgzNCwiaWF0IjoxNjE0ODc3NjQxLCJqdGkiOiIwYTU2ZmU5My05YzU1LTRhZWYtYjRjZS1mMTg3YTcwMWYwYTYifQ.MEQQrmnNfoNTxVT1ZOiFT-UDGqzbv12u6jAlwPvKjSaaJYV84QKU3NTU79mp9hdHY3J7sKgfTGGJ1qMJtfSFpg'

❗ IMPORTANT: Never push a secret hardcoded and in plaintext in a Git repository. This application is just for showcasing the use of JSON Web Tokens in the Petclinic application but it should never be done this way for real applications.

List of valid JSON Web Tokens:

Role: OWNER_ADMIN

eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJYMDAwMDEiLCJncm91cHMiOlsiUk9MRV9PV05FUl9BRE1JTiJdLCJleHAiOjQ3NzA1NTUyMTEsImlhdCI6MTYxNDg3ODAxOCwianRpIjoiMzU0NzZkNWEtMjA1OC00YzliLTlhMDMtMzkxZmJkOTgyMWU3In0.qwMEcq7l3m4ppazuPpVKGhm4q1KOeNQeK_wC09iUjvs1wngFszU8fwhBVa4DuwHuVTxgo2WGGLZCr5XsRc3kgA

Role: VET_ADMIN

eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJYMDAwMDIiLCJncm91cHMiOlsiUk9MRV9WRVRfQURNSU4iXSwiZXhwIjo0NzcwNTU1MjExLCJpYXQiOjE2MTQ4NzgwMTgsImp0aSI6ImU5ODNhMzY0LTNhMjMtNDRlMi05YTE0LTE4ZTNmYWQ2ZGRiZCJ9.8EWFF19GhHtjvIxQENnYkgLeCHje5RY2Jn67JyeErQDLsK6lZxspzmF7Eb2v-D1b_abf7Hxx0fWhtjVLQMEPPA

Role: ADMIN

eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJYMDAwMDMiLCJncm91cHMiOlsiUk9MRV9BRE1JTiJdLCJleHAiOjQ3NzA1NTUyMTEsImlhdCI6MTYxNDg3ODAxOCwianRpIjoiOTU1YWVlOGEtYzFkYi00YmJkLTg0ZGMtYjQwMmZkOWI2OTc3In0.4iXV2RBTwT35CkzPwOz2w1q54ts0_ozsyOE980tMpWcTJVBQw7bJBl2Y7Z_Dc-heUbI4VvBlTLGEhyBZIN8B-g

Roles: OWNER_ADMIN, VET_ADMIN, ADMIN

eyJhbGciOiJIUzUxMiJ9.eyJzdWIiOiJYMDAwMDQiLCJncm91cHMiOlsiUk9MRV9PV05FUl9BRE1JTiIsIlJPTEVfVkVUX0FETUlOIiwiUk9MRV9BRE1JTiJdLCJleHAiOjQ3NzA1NTUyMTEsImlhdCI6MTYxNDg3ODAxOCwianRpIjoiNGVmMjY0MjctYWUyNi00MGJjLWEzYjYtYzk3NGE0YTc0YWEzIn0.ypi1TQiGEuWqTKZwNe1pqMn7aIGXPCgMfONEoYmL2mrYW66vF5B3wFxVHyApcOOqttqNWtxNw7kPCdmqtW730Q

GitHub Actions CI/CD configuration

This section explains the necessary setup details to build and deploy the quarkus-petclinic-rest application to Google Cloud Platform (GCP) and Google Kubernetes Engine where we use Knative for serverless deployments.

Service Accounts

GitHub Actions is building the container images with Google Cloud Build and stores the resulting container image in Google Container Registry. Afterwards the image is going to be deployed to Google Cloud Run and Google Kubernetes Engine (GKE). To do this with a CI/CD tool like GitHub Actions we need to create Service Accounts to manage and regulate access to Google Cloud ressources.

Prerequisites

Before you start you should have already set up a GCP account with a billing account as well as a project. There are several ways to set up a service account with GCP. You can either use the Google Cloud SDK or the Management Console in your browser to create and configure service accounts.

Create a Service Account with gcloud

1. Export these environment variables so that you can copy and paste the following commands:

   export PROJECT_ID=<YOUR PROJECT ID>
   export SERVICE_ACCOUNT_NAME=<ENTER A NAME FOR YOUR SERVICE ACCOUNT>

2. Log in with your Google account:

   gcloud auth login

3. Select the project configured via $PROJECT_ID:

   gcloud config set project $PROJECT_ID

4. Enable the necessary services:

   gcloud services enable cloudbuild.googleapis.com run.googleapis.com containerregistry.googleapis.com container.googleapis.com

5. Create a service account:

   gcloud iam service-accounts create $SERVICE_ACCOUNT_NAME \
     --description="GitHub Actions service account for Petclinic repositories" \
     --display-name="$SERVICE_ACCOUNT_NAME"

6. Give the service account Cloud Run Admin, Storage Admin, and Service Account User roles. You can’t set all of them at once, so you have to run separate commands:

   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/cloudbuild.builds.editor
   
   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/iam.serviceAccountUser
   
   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/run.admin
    
   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/viewer
     
   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/storage.admin
     
   gcloud projects add-iam-policy-binding $PROJECT_ID \
     --member=serviceAccount:$SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com \
     --role=roles/container.admin  

7. Generate a key.json file with your credentials, so your GitHub workflow can authenticate with Google Cloud. After issuing the following command you can find the generated key in your current folder.

   gcloud iam service-accounts keys create key.json \
       --iam-account $SERVICE_ACCOUNT_NAME@$PROJECT_ID.iam.gserviceaccount.com

   The following tutorial explains this in more detail: link

Create a Service Account with GCP Management Console

1. Open the Google Cloud Management Console in your browser

2. Select your project with the dropdown on the top navigation bar

   

3. Go to IAM & Admin > Service Accounts and click ADDto create a new service account

4. Enter a name and a description (optional) for your new service account and click CREATE

   

5. Add the following roles to your service account and click DONE. The third step is not necessary and can be skipped.

   • Cloud Build Editor (roles/cloudbuild.builds.editor) - Required for Cloud Build
   • Service Account User (roles/iam.serviceAccountUser) - General Service Account permissions
   • Cloud Run Admin (roles/run.admin) - Required for Cloud Run
   • Viewer (roles/viewer) - Required as a workaround for successful deployments in GitHub Actions (see explanation below)
   • Storage Admin (roles/storage.admin) - Required for Container Registry
   • Kubernetes Engine Admin (roles/container.admin) - Required for GKE deployments

   Explanation of the Viewer Role:

   Link: https://towardsdatascience.com/deploy-to-google-cloud-run-using-github-actions-590ecf957af0 Once the service account is created you will need to select the following roles. I tried a number of different ways to remove the very permissive project viewer role, but at the time of this writing this your service account will need this role or the deployment will appear to fail in Github even if it is successfully deployed to Cloud Run.

   

6. Now click on your newly created service account and click ADD KEY on the service account details page. This will create new credentials which will be later used to authenticate your service account. Select JSONand click CREATE to generate and download your service account credentials.

   

7. Your service account credentials have been generated and downloaded on your machine. Make sure to keep them safe!

   

Setup Google Kubernetes Engine

1. Export these environment variables so that you can copy and paste the following commands:

   export PROJECT_ID=<YOUR PROJECT ID>
   export SERVICE_ACCOUNT_NAME=<ENTER A NAME FOR YOUR SERVICE ACCOUNT>

2. Log in with your Google account:

   gcloud auth login

3. Select the project configured via $PROJECT_ID:

   gcloud config set project $PROJECT_ID

4. Enable the necessary services (if not already enabled):

   gcloud services enable container.googleapis.com

5. Create GKE Cluster

   gcloud beta container --project "$PROJECT_ID" clusters create "petclinic-cluster" \
   --zone "europe-west3-c" \
   --no-enable-basic-auth \
   --cluster-version "1.18.12-gke.1210" \
   --release-channel "regular" \
   --machine-type "n2-standard-4" \
   --image-type "COS" \
   --disk-type "pd-standard" \
   --disk-size "100" \
   --metadata disable-legacy-endpoints=true \
   --scopes "https://www.googleapis.com/auth/devstorage.read_only","https://www.googleapis.com/auth/logging.write","https://www.googleapis.com/auth/monitoring","https://www.googleapis.com/auth/servicecontrol","https://www.googleapis.com/auth/service.management.readonly","https://www.googleapis.com/auth/trace.append" \
   --num-nodes "1" \
   --enable-stackdriver-kubernetes \
   --enable-ip-alias \
   --network "projects/$PROJECT_ID/global/networks/default" \
   --subnetwork "projects/$PROJECT_ID/regions/europe-west3/subnetworks/default" \
   --default-max-pods-per-node "110" \
   --no-enable-master-authorized-networks \
   --addons HorizontalPodAutoscaling,HttpLoadBalancing,GcePersistentDiskCsiDriver \
   --enable-autoupgrade \
   --enable-autorepair \
   --max-surge-upgrade 1 \
   --max-unavailable-upgrade 0 \
   --enable-shielded-nodes \
   --node-locations "europe-west3-c"

   After the cluster was successfuly created you should see the following output:

   NAME               LOCATION        MASTER_VERSION    MASTER_IP      MACHINE_TYPE   NODE_VERSION      NUM_NODES  STATUS
   petclinic-cluster  europe-west3-c  1.18.12-gke.1210  X.X.X.X        n2-standard-4  1.18.12-gke.1210  1          RUNNING
   

6. Connect to your Kubernetes Cluster

   gcloud container clusters get-credentials petclinic-cluster --region europe-west3 --project $PROJECT_ID

7. Create namespaces for the Petclinic application

   Create a Knative namespace for native executables on dev stage

   kubectl create namespace petclinic-native-knative-dev

   Create a Knative namespace for native images on prod stage

   kubectl create namespace petclinic-native-knative-prod

   Create a Knative namespace for JVM images on dev stage

   kubectl create namespace petclinic-jvm-knative-dev

   Create a Knative namespace for JVM images on prod stage

   kubectl create namespace petclinic-jvm-knative-prod

Setup Knative

1. Install the service component

   kubectl apply --filename https://github.com/knative/serving/releases/download/v0.21.0/serving-crds.yaml

2. Install the core components of Serving

   kubectl apply --filename https://github.com/knative/serving/releases/download/v0.21.0/serving-core.yaml

3. Install a properly configured Istio

   kubectl apply --filename https://github.com/knative/net-istio/releases/download/v0.21.0/istio.yaml

4. Install the Knative Istio controller

   kubectl apply --filename https://github.com/knative/net-istio/releases/download/v0.21.0/net-istio.yaml

5. Configure DNS with MagicDNS (xip.io)

   kubectl apply --filename https://github.com/knative/serving/releases/download/v0.21.0/serving-default-domain.yaml

   You might have to wait a few minutes and retry it later if you receive a connection time out error.

6. Monitor the Knative components until all of the components show a STATUS of Running or `Completed:

   kubectl get pods --namespace knative-serving

   Common Knative Commands:

   Show details about all Knative Services:

   kubectl get ksvc --all-namespaces

   Show details about quarkus-petclinic-rest application

   kubectl get ksvc quarkus-petclinic-rest --namespace <CHOOSE NAMESPACE FROM LIST BELOW>

   Available Namespaces:

   • petclinic-native-knative-dev - Namespace for Petclinic development containers running as a native executable
   • petclinic-native-knative-prod - Namespace for Petclinic production containers running as a native executable
   • petclinic-native-knative-dev - Namespace for Petclinic development containers running with a Java Virtual Machine (JVM)
   • petclinic-native-knative-prod - Namespace for Petclinic production containers running with a Java Virtual Machine (JVM)

GitHub configuration

In GitHub, you need to set up the following secrets via your repositories settings tab:

• GCP_PROJECT_ID - The GCP project id which was defined in $PROJECT_ID during the service account creation step
• GCP_SERVICE_ACCOUNT_EMAIL - The email from the previously created service account
• GCP_SERVICE_ACCOUNT_CREDENTIALS - The content from the key.json file that was previously created
• GCP_DB_HOST - The hostname or public IP address of the database server
• GCP_DB_PORT - The port of the database server
• GCP_DB_DATABASE - The database of your db server you are using for the application
• GCP_DB_USERNAME - The db username for your application
• GCP_DB_PASSWORD - The db user password for your application

In the end your secrets should look like this:

Google Cloud Run Endpoints - Development

Application	URL
Spring Petclinic (JVM)	https://spring-petclinic-rest-slim-jvm-dev-s2xflp6dzq-ey.a.run.app/petclinic/api

Google Cloud Run Endpoints - Production

Application	URL
Spring Petclinic (JVM)	https://spring-petclinic-rest-slim-jvm-prod-s2xflp6dzq-ey.a.run.app/petclinic/api