page_type: sample languages:
- java products:
- Azure Kubernetes Service description: "Deploy Spring Boot apps using AKS & MySQL" urlFragment: "spring-petclinic-microservices"
This microservices branch was initially derived from AngularJS version to demonstrate how to split sample Spring application into microservices. To achieve that goal we use IaC with Azure Bicep, MS build of OpenJDK 11, GitHub Actions, Azure AD Workload Identity, Azure Key Vault, Azure Container Registry, Azure Database for MySQL
See :
- the AKS Micro-services Reference Architecture
- https://github.com/Azure-Samples/java-on-azure-examples/tree/main/aks/springboot
To get an Azure subscription:
- If you have a Visual studio subscription then you can activate your free credits here
- If you do not currently have one, you can sign up for a free trial subscription here
To install Azure Bicep locally, read https://learn.microsoft.com/en-us/azure/azure-resource-manager/bicep/install
About how to build the container image, read :
Read :
- https://docs.github.com/en/actions/using-workflows/storing-workflow-data-as-artifacts
- https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-java-with-maven
- Overview of federated identity credentials in Azure Active Directory
You have to specify some KV secrets that will be then created in the GitHub Action Azure Infra services deployment workflow :
- SPRING-DATASOURCE-PASSWORD
- SPRING-CLOUD-AZURE-TENANT-ID
- VM-ADMIN-PASSWORD
dash '-' are not supported in GH secrets, so the secrets must be named in GH with underscore '_'.
(Also the '&' character in the SPRING_DATASOURCE_URL must be escaped with '&' jdbc:mysql://petcliaks777.mysql.database.azure.com:3306/petclinic?useSSL=true&requireSSL=true&enabledTLSProtocols=TLSv1.2&verifyServerCertificate=true)
Add the App secrets used by the Spring Config to your GH repo secrets / Actions secrets / Repository secrets / Add :
Secret Name | Secret Value example |
---|---|
SPRING_DATASOURCE_PASSWORD | PUT YOUR PASSWORD HERE |
SPRING_CLOUD_AZURE_TENANT_ID | PUT YOUR AZURE TENANT ID HERE |
VM_ADMIN_PASSWORD | PUT YOUR PASSWORD HERE |
LOCATION="westeurope"
RG_KV="rg-iac-kv33"
RG_APP="rg-iac-aks-petclinic-mic-srv"
az group create --name $RG_KV --location $LOCATION
az group create --name $RG_APP --location $LOCATION
A Service Principal is required for GitHub Action Runner, read https://aka.ms/azadsp-cli
SPN_APP_NAME="gha_aks_run"
# /!\ In CloudShell, the default subscription is not always the one you thought ...
subName="set here the name of your subscription"
subName=$(az account list --query "[?name=='${subName}'].{name:name}" --output tsv)
echo "subscription Name :" $subName
SUBSCRIPTION_ID=$(az account list --query "[?name=='${subName}'].{id:id}" --output tsv)
SUBSCRIPTION_ID=$(az account show --query id -o tsv)
TENANT_ID=$(az account show --query tenantId -o tsv)
Add your AZURE_SUBSCRIPTION_ID, AZURE_TENANT_ID to your GH repo Settings / Security / Secrets and variables / Actions / Actions secrets / Repository secrets
Read :
- https://learn.microsoft.com/en-us/azure/developer/github/connect-from-azure
- https://learn.microsoft.com/en-us/azure/developer/github/connect-from-azure?tabs=azure-cli%2Cwindows#use-the-azure-login-action-with-openid-connect
- https://learn.microsoft.com/en-us/azure/active-directory/develop/workload-identity-federation-create-trust?pivots=identity-wif-apps-methods-azp
In the GitHub Action Runner, to allow the Service Principal used to access the Key Vault, execute the command below:
#az ad app create --display-name $SPN_APP_NAME > aad_app.json
# This command will output JSON with an appId that is your client-id. The objectId is APPLICATION-OBJECT-ID and it will be used for creating federated credentials with Graph API calls.
#export APPLICATION_ID=$(cat aad_app.json | jq -r '.appId')
#export APPLICATION_OBJECT_ID=$(cat aad_app.json | jq -r '.id')
#az ad sp create --id $APPLICATION_ID
#export CREDENTIAL_NAME="gha_aks_run"
#export SUBJECT="repo:ezYakaEagle442/aks-java-petclinic-mic-srv:environment:PoC" # "repo:organization/repository:environment:Production"
#export DESCRIPTION="GitHub Action Runner for Petclinic AKS demo"
#az rest --method POST --uri 'https://graph.microsoft.com/beta/applications/$APPLICATION_OBJECT_ID/federatedIdentityCredentials' --body '{"name":"$CREDENTIAL_NAME","issuer":"https://token.actions.githubusercontent.com","subject":"$SUBJECT","description":"$DESCRIPTION","audiences":["api://AzureADTokenExchange"]}'
# SPN_PWD=$(az ad sp create-for-rbac --name $SPN_APP_NAME --skip-assignment --query password --output tsv)
az ad sp create-for-rbac --name $SPN_APP_NAME --skip-assignment --sdk-auth
{
"clientId": "<GUID>",
"clientSecret": "<GUID>",
"subscriptionId": "<GUID>",
"tenantId": "<GUID>",
"activeDirectoryEndpointUrl": "https://login.microsoftonline.com",
"resourceManagerEndpointUrl": "https://management.azure.com/",
"activeDirectoryGraphResourceId": "https://graph.windows.net/",
"sqlManagementEndpointUrl": "https://management.core.windows.net:8443/",
"galleryEndpointUrl": "https://gallery.azure.com/",
"managementEndpointUrl": "https://management.core.windows.net/"
}
Troubleshoot: If you hit "Error: : No subscriptions found for ***." , this is related to an IAM privilege in the subscription.
SPN_APP_ID=$(az ad sp list --all --query "[?appDisplayName=='${SPN_APP_NAME}'].{appId:appId}" --output tsv)
#SPN_APP_ID=$(az ad sp list --show-mine --query "[?appDisplayName=='${SPN_APP_NAME}'].{appId:appId}" --output tsv)
# TENANT_ID=$(az ad sp list --show-mine --query "[?appDisplayName=='${SPN_APP_NAME}'].{t:appOwnerOrganizationId}" --output tsv)
# Enterprise Application
az ad app list --show-mine --query "[?displayName=='${SPN_APP_NAME}'].{objectId:id}"
az ad app show --id $SPN_APP_ID
# This is the unique ID of the Service Principal object associated with this application.
# SPN_OBJECT_ID=$(az ad sp list --show-mine --query "[?displayName=='${SPN_APP_NAME}'].{objectId:id}" -o tsv)
SPN_OBJECT_ID=$(az ad sp list --all --query "[?displayName=='${SPN_APP_NAME}'].{objectId:id}" -o tsv)
az ad sp show --id $SPN_OBJECT_ID
# the assignee is an appId
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role contributor
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role contributor
# https://learn.microsoft.com/en-us/azure/key-vault/general/rbac-guide?tabs=azure-cli#azure-built-in-roles-for-key-vault-data-plane-operations
# "Key Vault Secrets User"
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role 4633458b-17de-408a-b874-0445c86b69e6
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role 4633458b-17de-408a-b874-0445c86b69e6
# "Key Vault Secrets Officer"
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role b86a8fe4-44ce-4948-aee5-eccb2c155cd7
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role b86a8fe4-44ce-4948-aee5-eccb2c155cd7
# "DNS Zone Contributor"
# https://learn.microsoft.com/en-us/azure/role-based-access-control/built-in-roles#dns-zone-contributor
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role befefa01-2a29-4197-83a8-272ff33ce314
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role befefa01-2a29-4197-83a8-272ff33ce314
# https://learn.microsoft.com/en-us/azure/role-based-access-control/built-in-roles#virtual-machine-contributor
# Virtual Machine Contributor has permission 'Microsoft.Network/publicIPAddresses/read'
#az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role 9980e02c-c2be-4d73-94e8-173b1dc7cf3c
#az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role 9980e02c-c2be-4d73-94e8-173b1dc7cf3c
# Network-contributor: https://learn.microsoft.com/en-us/azure/role-based-access-control/resource-provider-operations#microsoftnetwork
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role 4d97b98b-1d4f-4787-a291-c67834d212e7
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID} --role 4d97b98b-1d4f-4787-a291-c67834d212e7
# https://learn.microsoft.com/en-us/azure/role-based-access-control/role-assignments-portal#prerequisites
# /!\ To assign Azure roles, you must have: requires to have Microsoft.Authorization/roleAssignments/write and Microsoft.Authorization/roleAssignments/delete permissions,
# such as User Access Administrator or Owner.
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role Owner
az role assignment create --assignee $SPN_APP_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_APP} --role Owner
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_KV} --role Owner
az role assignment create --assignee $SPN_OBJECT_ID --scope /subscriptions/${SUBSCRIPTION_ID}/resourceGroups/${RG_APP} --role Owner
**RBAC Permission model is set on KV, the pre-req requires to have Microsoft.Authorization/roleAssignments/write and Microsoft.Authorization/roleAssignments/delete permissions, such as User Access Administrator or Owner.
https://learn.microsoft.com/en-us/azure/role-based-access-control/role-assignments-portal#prerequisites To assign Azure roles, you must have: requires to have Microsoft.Authorization/roleAssignments/write and Microsoft.Authorization/roleAssignments/delete permissions, such as User Access Administrator or Owner. **
"Key Vault Secrets User" built-in role read secret contents including secret portion of a certificate with private key. Only works for key vaults that use the 'Azure role-based access control' permission model.
Read :
- Use GitHub Actions to connect to Azure documentation.
- https://github.com/Azure/login#configure-a-service-principal-with-a-secret
Paste in your JSON object for your service principal with the name AZURE_CREDENTIALS as secrets to your GH repo Settings / Security / Secrets and variables / Actions / Actions secrets / Repository secrets
You can test your connection with CLI :
az login --service-principal -u $SPN_APP_ID -p $SPN_PWD --tenant $TENANT_ID
Add SUBSCRIPTION_ID, TENANT_ID, SPN_APP_ID and SPN_PWD as secrets to your GH repo Settings / Security / Secrets and variables / Actions / Actions secrets / Repository secrets
Finally Create a GH PAT "PKG_PAT" that can be use to , publish packages and delete packages
Your GitHub personal access token needs to have the workflow scope selected. You need at least delete:packages and read:packages scopes to delete a package. You need contents: read and packages: write permissions to publish and download artifacts
Create SSH Keys, WITHOUT any passphrase (type enter if prompt)
# https://github.com/Azure/azure-quickstart-templates/blob/master/quickstarts/microsoft.resources/deployment-script-ssh-key-gen/new-key.sh
export ssh_key=aksadm
echo -e 'y' | ssh-keygen -t rsa -b 4096 -f ~/.ssh/$ssh_key -C "[email protected]" # -N $ssh_passphrase
# test
# ssh -i ~/.ssh/$ssh_key $admin_username@$network_interface_pub_ip
Add $ssh_key & $ssh_key.pub as secrets SSH_PRV_KEY & SSH_PUB_KEY to your GH repo Settings / Security / Secrets and variables / Actions / Actions secrets / Repository secrets
To avoid to hit the error below :
"The subscription is not registered to use namespace 'Microsoft.KeyVault'. See https://aka.ms/rps-not-found for how to register subscriptions.\",\r\n \"details\": [\r\n ***\r\n \"code\": \"MissingSubscriptionRegistration\"
Read the docs Just run :
az feature list --output table --namespace Microsoft.ContainerService
az feature register --namespace "Microsoft.ContainerService" --name "AKS-GitOps"
az feature register --namespace "Microsoft.ContainerService" --name "EnableWorkloadIdentityPreview"
az feature register --namespace "Microsoft.ContainerService" --name "AKS-Dapr"
az feature register --namespace "Microsoft.ContainerService" --name "EnableAzureKeyvaultSecretsProvider"
az feature register --namespace "Microsoft.ContainerService" --name "AKS-AzureDefender"
az feature register --namespace "Microsoft.ContainerService" --name "AKS-PrometheusAddonPreview"
az feature register --namespace "Microsoft.ContainerService" --name "AutoUpgradePreview"
az feature register --namespace "Microsoft.ContainerService" --name "AKS-OMSAppMonitoring"
az feature register --namespace "Microsoft.ContainerService" --name "ManagedCluster"
az feature register --namespace "Microsoft.ContainerService" --name "AKS-AzurePolicyAutoApprove"
az feature register --namespace "Microsoft.ContainerService" --name "FleetResourcePreview"
az provider list --output table
az provider list --query "[?registrationState=='Registered']" --output table
az provider list --query "[?namespace=='Microsoft.KeyVault']" --output table
az provider list --query "[?namespace=='Microsoft.OperationsManagement']" --output table
az provider register --namespace Microsoft.KeyVault
az provider register --namespace Microsoft.ContainerRegistry
az provider register --namespace Microsoft.ContainerService
az provider register --namespace Microsoft.OperationalInsights
az provider register --namespace Microsoft.DBforMySQL
az provider register --namespace Microsoft.DBforPostgreSQL
az provider register --namespace Microsoft.Compute
az provider register --namespace Microsoft.AppConfiguration
az provider register --namespace Microsoft.AppPlatform
az provider register --namespace Microsoft.EventHub
az provider register --namespace Microsoft.Kubernetes
az provider register --namespace Microsoft.KubernetesConfiguration
az provider register --namespace Microsoft.Kusto
az provider register --namespace Microsoft.ManagedIdentity
az provider register --namespace Microsoft.Monitor
az provider register --namespace Microsoft.OperationsManagement
az provider register --namespace Microsoft.Network
c
az provider register --namespace Microsoft.ServiceBus
az provider register --namespace Microsoft.Storage
az provider register --namespace Microsoft.Subscription
# https://learn.microsoft.com/en-us/azure/aks/cluster-extensions
az extension add --name k8s-extension
az extension update --name k8s-extension
# https://learn.microsoft.com/en-us/azure/azure-arc/kubernetes/tutorial-use-gitops-flux2?
az extension add -n k8s-configuration
Read https://azure.github.io/azure-workload-identity/docs/installation/azwi.html
Install Azure AD Workload Identity CLI
AAD_WI_CLI_VERSION=1.0.0
wget https://github.com/Azure/azure-workload-identity/releases/download/v$AAD_WI_CLI_VERSION/azwi-v$AAD_WI_CLI_VERSION-linux-amd64.tar.gz
gunzip azwi-v$AAD_WI_CLI_VERSION-linux-amd64.tar.gz
tar -xvf azwi-v$AAD_WI_CLI_VERSION-linux-amd64.tar
./azwi version
See GitHub Actions :
- Deploy the Azure Infra services workflow
- Maven Backends Build workflow
- Maven UI Build workflow
- Java Apps Backends Deploy workflow
- Java Apps UI Deploy workflow
- Delete ALL the Azure Infra services workflow, except KeyVault
****
Workflow Design
The Workflow run the steps in this in this order :
├── Deploy the Azure Infra services workflow ./.github/workflows/deploy-iac.yml
│ ├── Trigger the pre-req ./.github/workflows/deploy-iac.yml#L75
│ ├── Create Azure Key Vault ./.github/workflows/deploy-iac-pre-req.yml#L108
│ ├── Authorize local IP to access the Azure Key Vault ./.github/workflows/deploy-iac-pre-req.yml#L115
│ ├── Create the secrets ./.github/workflows/deploy-iac-pre-req.yml#L121
│ ├── Disable local IP access to the Key Vault ./.github/workflows/deploy-iac-pre-req.yml#L152
│ ├── Deploy the pre-req ./.github/workflows/deploy-iac-pre-req.yml#L180
│ ├── Create Log Analytics Workspace ./iac/bicep/pre-req.bicep#L68
│ ├── Create appInsights ./iac/bicep/pre-req.bicep#L68
│ ├── Create ACR ./iac/bicep/pre-req.bicep#L104
│ ├── Create Identities ./iac/bicep/pre-req.bicep#L124
│ ├── Create VNet ./iac/bicep/pre-req.bicep#L135
│ ├── Create roleAssignments ./iac/bicep/pre-req.bicep#L155
│ ├── Create MySQL ./iac/bicep/pre-req.bicep#L174
│ ├── Deploy AKS ./iac/bicep/main.bicep
│ ├── Call AKS module ./iac/bicep/main.bicep#L95
│ ├── Whitelist AKS Env. OutboundIP to KV and MySQL ./.github/workflows/deploy-iac.yml#L119
│ ├── Call DB data loading Init ./.github/workflows/deploy-iac.yml#L154
│ ├── Call Maven Build ./.github/workflows/deploy-iac.yml#L159
│ ├── Maven Build ./.github/workflows/maven-build.yml#L128
│ ├── Publish the Maven package ./.github/workflows/maven-build.yml#L176
│ ├── Build image and push it to ACR ./.github/workflows/maven-build.yml#L241
│ ├── Call Maven Build-UI ./.github/workflows/deploy-iac.yml#L166
│ ├── Build image and push it to ACR ./.github/workflows/maven-build-ui.yml#L191
│ ├── Deploy Backend Services ./.github/workflows/deploy-iac.yml#L185
│ ├── Deploy Backend services calling ./.github/workflows/deploy-app-svc.yml
│ ├── Deploy the UI calling ./.github/workflows/deploy-app-ui.yml
You need to set your own param values in :
env:
APP_NAME: petcliaks
LOCATION: westeurope # francecentral
RG_KV: rg-iac-kv33 # RG where to deploy KV
RG_APP: rg-iac-aks-petclinic-mic-srv # RG where to deploy the other Azure services: AKS, ACR, MySQL, etc.
ACR_NAME: acrpetcliaks
VNET_NAME: vnet-aks
VNET_CIDR: 172.16.0.0/16
AKS_SUBNET_CIDR: 172.16.1.0/24
AKS_SUBNET_NAME: snet-aks
START_IP_ADRESS: 172.16.1.0
END_IP_ADRESS: 172.16.1.255
MYSQL_SERVER_NAME: petcliaks
MYSQL_DB_NAME: petclinic
MYSQL_ADM_USR: mys_adm
MYSQL_TIME_ZONE: Europe/Paris
MYSQL_CHARACTER_SET: utf8
MYSQL_PORT: 3306
DEPLOY_TO_VNET: false
KV_NAME: kv-petcliaks33 # The name of the KV, must be UNIQUE. A vault name must be between 3-24 alphanumeric characters
# https://learn.microsoft.com/en-us/azure/key-vault/secrets/secrets-best-practices#secrets-rotation
# Because secrets are sensitive to leakage or exposure, it's important to rotate them often, at least every 60 days.
# Expiry date in seconds since 1970-01-01T00:00:00Z. Ex: 1672444800 ==> 31/12/2022'
SECRET_EXPIRY_DATE: 1703980800 # ==> 31/12/2023
AZURE_CONTAINER_REGISTRY: acrpetcliaks # The name of the ACR, must be UNIQUE. The name must contain only alphanumeric characters, be globally unique, and between 5 and 50 characters in length.
REGISTRY_URL: acrpetcliaks.azurecr.io # set this to the URL of your registry
REPOSITORY: petclinic # set this to your ACR repository
PROJECT_NAME: petclinic # set this to your project's name
KV_NAME: kv-petcliaks33 # The name of the KV, must be UNIQUE. A vault name must be between 3-24 alphanumeric characters
RG_KV: rg-iac-kv33 # RG where to deploy KV
RG_APP: rg-iac-aks-petclinic-mic-srv # RG where to deploy the other Azure services: AKS, ACR, MySQL, etc.
# ==== Azure storage to store Artifacts , values must be consistent with the ones in storage.bicep ====:
AZ_STORAGE_NAME : stakspetcliaks # customize this
AZ_BLOB_CONTAINER_NAME: petcliaks-blob # customize this
AZURE_CONTAINER_REGISTRY: acrpetcliaks # The name of the ACR, must be UNIQUE. The name must contain only alphanumeric characters, be globally unique, and between 5 and 50 characters in length.
REGISTRY_URL: acrpetcliaks.azurecr.io # set this to the URL of your registry
REPOSITORY: petclinic # set this to your ACR repository
PROJECT_NAME: petclinic # set this to your project's name
KV_NAME: kv-petcliaks33 # The name of the KV, must be UNIQUE. A vault name must be between 3-24 alphanumeric characters
RG_KV: rg-iac-kv33 # RG where to deploy KV
RG_APP: rg-iac-aks-petclinic-mic-srv # RG where to deploy the other Azure services: AKS, ACR, MySQL, etc.
# ==== Azure storage to store Artifacts , values must be consistent with the ones in storage.bicep ====:
AZ_STORAGE_NAME : stakspetcliaks # customize this
AZ_BLOB_CONTAINER_NAME: petcliaks-blob # customize this
Once you commit, then push your code update to your repo, it will trigger a Maven build which you need to can CANCELL from https://github.com/USERNAME/aks-java-petclinic-mic-srv/actions/workflows/maven-build.yml the first time you trigger the workflow, anyway it will fail because the ACR does not exist yet and the docker build will fail to push the Images.
Note: the GH Hosted Runner / Ubuntu latest image has already Azure CLI installed
You can read the Bicep section but you do not have to run it through CLI, instead you can manually trigger the GitHub Action deploy-iac.yml, see the Workflow in the next section
AKS has dependencies on services outside of that virtual network. For a list of these dependencies see the AKS doc
Troubleshoot : If the AKS cluster was provisionned in a FAILED state, try :
az resource update --name $ClusterName --resource-group $RgName --resource-type Microsoft.ContainerService/managedClusters --debug
az resource show --name $ClusterName --resource-group $RgName --resource-type Microsoft.ContainerService/managedClusters --debug
Azure Key Vault integration is implemented through Spring Cloud for Azure
Read :
- https://learn.microsoft.com/en-us/azure/developer/java/spring-framework/configure-spring-boot-starter-java-app-with-azure-key-vault
- [https://microsoft.github.io/spring-cloud-azure/current/reference/html/index.html#advanced-usage]https://microsoft.github.io/spring-cloud-azure/current/reference/html/index.html#advanced-usage)
- Azure/azure-sdk-for-java#28310
- Maven Project parent pom.xml
The Config-server does use the config declared on the repo at https://github.com/ezYakaEagle442/aks-cfg-srv/blob/main/application.yml and uses a User-Assigned Managed Identity to be able to read secrets from KeyVault.
If you face any issue, see the troubleshoot section
Quick local test just to verify that the jar files can be run (the routing will not work out of a K8S cluster, and also the apps will fail to start as soon as management port 8081 will be already in use by config server ...) :
/!\ IMPORTANT WARNING: projects must be built with -Denv=cloud EXCEPT for api-gateway
mvn clean package -DskipTests -Denv=azure
java -jar spring-petclinic-config-server\target\spring-petclinic-config-server-2.6.13.jar --server.port=8888
java -jar spring-petclinic-admin-server\target\spring-petclinic-admin-server-2.6.13.jar --server.port=9090
java -jar spring-petclinic-visits-service\target\spring-petclinic-visits-service-2.6.13.jar --server.port=8082 # --spring.profiles.active=docker
java -jar spring-petclinic-vets-service\target\spring-petclinic-vets-service-2.6.13.jar --server.port=8083
java -jar spring-petclinic-customers-service\target\spring-petclinic-customers-service-2.6.13.jar --server.port=8084
java -jar spring-petclinic-api-gateway\target\spring-petclinic-api-gateway-2.6.13.jar --server.port=8085
Note: tip to verify the dependencies
mvn dependency:tree
mvn dependency:analyze-duplicate
To learn more about maven, read :
- https://www.baeldung.com/maven
- https://www.baeldung.com/maven-duplicate-dependencies
- https://www.baeldung.com/maven-multi-module
Every microservice is a Spring Boot application and can be started locally. Please note that supporting services (Config Server) must be started before any other application (Customers, Vets, Visits and API). Startup Admin server is optional. If everything goes well, you can access the following services at given location:
- AngularJS frontend (API Gateway) - http://localhost:8080
- Admin Server (Spring Boot Admin) - http://localhost:9090
The main
branch uses an MS openjdk/jdk:11-mariner Docker base.
#acr_usr=$(az deployment group show -g ${{ env.RG_APP }} -n ${{ env.AZURE_CONTAINER_REGISTRY }} --query properties.outputs.acrRegistryUsr.value | tr -d '"')
#acr_pwd=$(az deployment group show -g ${{ env.RG_APP }} -n ${{ env.AZURE_CONTAINER_REGISTRY }} --query properties.outputs.acrRegistryPwd.value | tr -d '"')
#az acr login --name ${{ env.REGISTRY_URL }} -u $acr_usr -p $acr_pwd
set -euo pipefail
access_token=$(az account get-access-token --query accessToken -o tsv)
refresh_token=$(curl https://${{ env.REGISTRY_URL }}/oauth2/exchange -v -d "grant_type=access_token&service=${{ env.REGISTRY_URL }}&access_token=$access_token" | jq -r .refresh_token)
refresh_token=$(curl https://acrpetcliaks.azurecr.io/oauth2/exchange -v -d "grant_type=access_token&service=acrpetcliaks.azurecr.io&access_token=$access_token" | jq -r .refresh_token)
# docker login ${{ env.REGISTRY_URL }} -u 00000000-0000-0000-0000-000000000000 --password-stdin <<< "$refresh_token"
docker build --build-arg --no-cache -t "petclinic-admin-server" -f "./docker/petclinic-admin-server/Dockerfile" .
docker tag petclinic-admin-server acrpetcliaks.azurecr.io/petclinic/petclinic-admin-server
az acr login --name acrpetcliaks.azurecr.io -u $acr_usr -p $acr_pwd
az acr build --registry acrpetcliaks -g rg-iac-aks-petclinic-mic-srv -t petclinic/adm-test:test --file "./docker/petclinic-admin-server/Dockerfile" .
docker push acrpetcliaks.azurecr.io/petclinic/petclinic-admin-server
docker pull acrpetcliaks.azurecr.io/petclinic/petclinic-admin-server
docker image ls
Note: the Docker files must be named Dockerfile See Azure/azure-cli-extensions#5041
See the presentation of the Spring Petclinic Framework version
A blog bost introducing the Spring Petclinic Microsevices (french language)
You can then access petclinic here: http://localhost:8080/
Architecture diagram of the Spring Petclinic Microservices
The UI code is located at spring-petclinic-api-gateway\src\main\resources\static\scripts
The Spring Zuul(Netflix Intelligent Routing) config at https://github.com/ezYakaEagle442/aks-cfg-srv/blob/main/api-gateway.yml has been deprecated and replaced by the Spring Cloud Gateway.
The Spring Cloud Gateway routing is configured at spring-petclinic-api-gateway/src/main/resources/application.yml
The API Gateway Controller is located at spring-petclinic-api-gateway/src/main/java/org/springframework/samples/petclinic/api/boundary/web/ApiGatewayController.java
Note: The Spring Cloud Discovery Server is NOT deployed as the underlying K8S/AKS discovery/DNS service is used. see :
- https://spring.io/guides/gs/service-registration-and-discovery
- https://spring.io/projects/spring-cloud-netflix
- [https://kubernetes.io/docs/tasks/administer-cluster/dns-debugging-resolutionhttps://kubernetes.io/docs/tasks/administer-cluster/dns-debugging-resolution/
- https://kubernetes.io/docs/concepts/services-networking/dns-pod-service/
The K8S routing is configured in the Ingress resources at :
- spring-petclinic-api-gateway\k8s\petclinic-ui-ingress.yaml
- spring-petclinic-admin-server\k8s\petclinic-admin-server-ingress.yaml
- spring-petclinic-config-server\k8s\petclinic-config-server-ingress.yaml
- spring-petclinic-customers-service\k8s\petclinic-customer-ingress.yaml
- spring-petclinic-vets-service\k8s\petclinic-vet-ingress.yaml
- spring-petclinic-visits-service\k8s\petclinic-visits-ingress.yaml
The Git repo URL used by Spring config is set in spring-petclinic-config-server/src/main/resources/application.yml
If you want to know more about the Spring Boot Admin server, you might be interested in https://github.com/codecentric/spring-boot-admin
For Learning purpose the App uses Key Vault to fetch secrets like the DB password but if would be even better using Passwordless Features: https://aka.ms/delete-passwords
Spring Boot is a framework aimed to help developers to easily create and build stand-alone, production-grade Spring based Applications that you can “just run”.
Spring Cloud Config provides server and client-side support for externalized configuration in a distributed system. With the Spring Cloud Config Server you have a central place to manage external properties for applications across all environments.
Spring Cloud Config Server is a centralized service that via HTTP provides all the applications configuration (name-value pairs or equivalent YAML content). The server is embeddable in a Spring Boot application, by using the @EnableConfigServer annotation.
In other words, the Spring Cloud Config Server is simply a Spring Boot application, configured as a Spring Cloud Config Server, and that is able to retrieve the properties from the configured property source. The property source can be a Git repository, svn or Consul service.
A Spring Boot application properly configured, can take immediate advantage of the Spring Config Server. It also picks up some additional useful features related to Environment change events. Any Spring Boot application can easily be configured as a Spring Cloud Config Client.
See the Azure doc Each micro-service is containerized using a Dockerfile. Example at ./docker/petclinic-customers-service/Dockerfile
About how to build the container image, read ACR doc
In its default configuration, Petclinic uses an in-memory database (HSQLDB) which gets populated at startup with data.
A similar setup is provided for MySql in case a persistent database configuration is needed.
Dependency for Connector/J, the MySQL JDBC driver is already included in the pom.xml
files.
You need to reconfigure the MySQL connection string with your own settings (you can get it from the Azure portal / petcliaks-mysql-server / Connection strings / JDBC): In the spring-petclinic-microservices-config/blob/main/application.yml :
spring:
config:
activate:
on-profile: mysql
datasource:
schema: classpath*:db/mysql/schema.sql
data: classpath*:db/mysql/data.sql
url: jdbc:mysql://petcliaks.mysql.database.azure.com:3306/petclinic?useSSL=true&requireSSL=true&enabledTLSProtocols=TLSv1.2&verifyServerCertificate=true
In fact the spring.datasource.password will be automatically injected from KV secrets SPRING-DATASOURCE-PASSWORD using the config below in each micro-service : example for Customers-Service spring-petclinic-customers-service/src/main/resources/application.yml
spring:
cloud:
azure:
profile: # spring.cloud.azure.profile
# subscription-id:
tenant-id: ${AZURE_TENANT_ID}
credential:
managed-identity-enabled: true
keyvault:
secret:
enabled: true
property-sources:
- name: kv-cfg-XXX # KV Config for each App XXX
endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
credential:
managed-identity-enabled: true
client-id: ${XXXX_SVC_APP_IDENTITY_CLIENT_ID}
---
You can check the DB connection with this sample project.
To use a MySQL database, you have to start 3 microservices (visits-service
, customers-service
and vets-services
)
with the mysql
Spring profile. Add the --spring.profiles.active=mysql
as programm argument.
In the application.yml
of the [Configuration repository], set the initialization-mode
to never
( or ALWAYS
).
If you are running the microservices with Docker, you have to add the mysql
profile into the (Dockerfile)[docker/Dockerfile]:
ENV SPRING_PROFILES_ACTIVE docker,mysql
All MySQL flexible-server parameters are set in the sql-load workflow called by the IaC deployment workflow
Read the Application Insights docs :
- https://learn.microsoft.com/en-us/azure/azure-monitor/app/java-in-process-agent
- https://learn.microsoft.com/en-us/azure/azure-monitor/app/java-spring-boot#spring-boot-via-docker-entry-point
- https://learn.microsoft.com/en-us/azure/azure-monitor/app/java-in-process-agent#set-the-application-insights-connection-string
- https://techcommunity.microsoft.com/t5/apps-on-azure-blog/bg-p/AppsonAzureBlog
- https://github.com/microsoft/ApplicationInsights-Java
- https://github.com/microsoft/AzureMonitorCommunity
- https://learn.microsoft.com/en-us/azure/azure-monitor/containers/container-insights-log-query
The config files are located in each micro-service at src/main/resources/applicationinsights.json The Java agent is downloaded in the App container in /tmp/app, you can have a look at a Docker file, example at ./docker/petclinic-customers-service/Dockerfile
You can specify your own configuration file path by using one of these two options:
- APPLICATIONINSIGHTS_CONFIGURATION_FILE environment variable
- applicationinsights.configuration.file Java system property
In our configuration, in the containers the applicationinsights.json is located at BOOT-INF/classes/applicationinsights.json so we must set APPLICATIONINSIGHTS_CONFIGURATION_FILE=BOOT-INF/classes/applicationinsights.json
Open the Petclinic application and try out a few tasks - view pet owners and their pets, vets, and schedule pet visits:
open http://petclinic.westeurope.cloudapp.azure.com//
You can also use your browser or curl
the REST API exposed by the Petclinic application.
The admin REST API allows you to create/update/remove items in Pet Owners, Pets, Vets and Visits.
You can run the following curl commands:
URL ex:
- https://..<cloudapp.azure.com>
- http://petclinic.westeurope.cloudapp.azure.com/owners/3/pets/4
- http://petclinic.westeurope.cloudapp.azure.com/owners/6/pets/8/visits
- http://petclinic.westeurope.cloudapp.azure.com/vets
with Custom domains : http://appinnohandsonlab.com/#!/welcome
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/owners
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/owners/4
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/owners/
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/petTypes
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/owners/3/pets/4
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/customer/owners/6/pets/8/
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/vet/vets
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/visit/owners/6/pets/8/visits
curl -X GET http://petclinic.westeurope.cloudapp.azure.com/api/visit/owners/6/pets/8/visits
Spring Boot includes a number of additional features to help you monitor and manage your application when you push it to production (Spring Boot Actuator: Production-ready Features). You can choose to manage and monitor your application by using HTTP endpoints or with JMX. Auditing, health, and metrics gathering can also be automatically applied to your application.
Actuator endpoints let you monitor and interact with your application. By default, Spring Boot application exposes health
and info
endpoints to show arbitrary application info and health information. Apps in this project are pre-configured to expose all the Actuator endpoints.
You can try them out by opening the following app actuator endpoints in a browser:
http://petclinic.westeurope.cloudapp.azure.com
open http://petclinic.westeurope.cloudapp.azure.com/manage/
open http://petclinic.westeurope.cloudapp.azure.com/manage/env
open http://petclinic.westeurope.cloudapp.azure.com/manage/configprops
open http://petclinic.westeurope.cloudapp.azure.com/api/customer/manage
open http://petclinic.westeurope.cloudapp.azure.com/api/customer/manage/env
open http://petclinic.westeurope.cloudapp.azure.com/api/customer/manage/configprops
### Monitor Petclinic logs and metrics in Azure Log Analytics
To get the App logs :
```bash
LOG_ANALYTICS_WORKSPACE_CLIENT_ID=`az monitor log-analytics workspace show -n $LOG_ANALYTICS_WORKSPACE -g $RESOURCE_GROUP --query customerId --out tsv`
az monitor log-analytics query \
--workspace $LOG_ANALYTICS_WORKSPACE_CLIENT_ID \
--analytics-query "ContainerLog | where LogEntry has 'error' |take 100" \
--out table
Open the Log Analytics that you created - you can find the Log Analytics in the same Resource Group where you created the AKS cluster.
In the Log Analyics page, selects Logs blade and run any of the sample queries supplied below for AKS.
Type and run the following Kusto query to see all the logs from the AKS Service :
// https://learn.microsoft.com/en-us/azure/azure-monitor/containers/container-insights-log-query
let startTimestamp = ago(1h);
KubePodInventory
| where TimeGenerated > startTimestamp
| project ContainerID, PodName=Name, Namespace
| where PodName contains "service" and Namespace startswith "petclinic"
| distinct ContainerID, PodName
| join
(
ContainerLog
| where TimeGenerated > startTimestamp
)
on ContainerID
// at this point before the next pipe, columns from both tables are available to be "projected". Due to both
// tables having a "Name" column, we assign an alias as PodName to one column which we actually want
| project TimeGenerated, PodName, LogEntrySource, LogEntry
| summarize by TimeGenerated, LogEntry
| order by TimeGenerated desc
let FindString = "error";//Please update term you would like to find in LogEntry here
ContainerLog
| where LogEntry has FindString
| take 100
Spring Boot registers a lot number of core metrics: JVM, CPU, Tomcat, Logback...
The Spring Boot auto-configuration enables the instrumentation of requests handled by Spring MVC.
All those three REST controllers OwnerResource
, PetResource
and VisitResource
have been instrumented by the @Timed
Micrometer annotation at class level.
customers-service
application has the following custom metrics enabled:- @Timed:
petclinic.owner
- @Timed:
petclinic.pet
- @Timed:
visits-service
application has the following custom metrics enabled:- @Timed:
petclinic.visit
- @Timed:
Circuit breakers TODO !
If you face this error :
Caused by: java.sql.SQLException: Connections using insecure transport are prohibited while --require_secure_transport=ON.
It might be related to the Spring Config configured at https://github.com/Azure-Samples/spring-petclinic-microservices-config/blob/master/application.yml which on-profile: mysql is set with datasource url : jdbc:mysql://${MYSQL_SERVER_FULL_NAME}:3306/${MYSQL_DATABASE_NAME}?useSSL=false
Check the MySQL connector doc Your JBCC URL should look like this for instance: url: jdbc:mysql://localhost:3306/petclinic?useSSL=false url: jdbc:mysql://${MYSQL_SERVER_FULL_NAME}:3306/${MYSQL_DATABASE_NAME}??useSSL=true url: jdbc:mysql://petclinic-mysql-server.mysql.database.azure.com:3306/petclinic?useSSL=true url: jdbc:mysql://petclinic-mysql-server.mysql.database.azure.com:3306/petclinic?useSSL=true&requireSSL=true&enabledTLSProtocols=TLSv1.2&verifyServerCertificate=true
If you face this Netty SSL Handshake issue :
eactor.core.Exceptions$ReactiveException: io.netty.handler.ssl.SslHandshakeTimeoutException: handshake timed out after 10000ms
It means that you may need to upgrade your Spring Boot version to the latest one... See netty/netty#12343
If you face this issue :
error Caused by: java.net.MalformedURLException: no protocol: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
It means that the api-gateway project had been built with mvn -B clean package --file pom.xml -DskipTests -Denv=cloud This set the env=cloud at in the parent POM which then injects the spring-cloud-azure-starter-keyvault-secrets dependency at POM it looks like event just having such dependency would cause the runtime to look for ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
If you face this issue :
Spring MVC found on classpath, which is incompatible with Spring Cloud Gateway
Please set spring.main.web-application-type=reactive or remove spring-boot-starter-web dependency.
See: -https://cloud.spring.io/spring-cloud-gateway/reference/html/#gateway-starter
spring-cloud-starter-netflix-eureka-server depends on spring-boot-starter-web you would need to remove the dependency on spring-boot-starter-web in the api-gateway module
check with : mvn dependency:tree
mvn dependency:tree | grep spring-boot-starter-web
About How to use Env. variable in Spring Boot, see :
- https://docs.spring.io/spring-boot/docs/current/reference/html/features.html#features.external-config
- https://www.baeldung.com/spring-boot-properties-env-variables
https://learn.microsoft.com/en-us/azure/spring-apps/tutorial-managed-identities-key-vault?tabs=user-assigned-managed-identity Fast-Track for Azure OpenLab aka Java OpenHack uses SYSTEM-Assigned MI
The Azure SDK API change is summarized at Issue #28310
KeyVault integration runs easily when :
- You use SYSTEM-Assigned MI, because then in the Config use by the Config-server you do NOT need to specify the client-id
- When you use 1 & only 1 USER-Assigned MI for ALL your Apps/Micro-services, this is not a good practice from a security perspective as it is safer to assign 1 Identity to each App
When you use USER-Assigned MI, assigning 1 Identity to each App , see one App in Bicep. In the Config used by the Config-server if you declare as many property-sources as the number of micro-services setting the client-id with the App Id (using Env. Var. set in the GH Workflow) :
keyvault:
secret:
enabled: true
property-source-enabled: true
property-sources:
- name: kv-cfg-vets # KV Config for each App Vets-Service
endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
credential:
managed-identity-enabled: true
client-id: ${VETS_SVC_APP_IDENTITY_CLIENT_ID}
# client-secret: ${AZURE_CLIENT_SECRET} for SPN not for MI
# profile:
# tenant-id: ${SPRING_CLOUD_AZURE_TENANT_ID}
- name: kv-cfg-visits # KV Config for each App Visits-Service
endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
credential:
managed-identity-enabled: true
client-id: ${VISITS_SVC_APP_IDENTITY_CLIENT_ID}
- name: kv-cfg-customers # KV Config for each App Customers-Service
endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
credential:
managed-identity-enabled: true
client-id: ${CUSTOMERS_SVC_APP_IDENTITY_CLIENT_ID}
As a consequence this initially failed as each App uses the above Config and tried to fetch KV secrets from other App property-sources. which failed because it was not allowed as it was assigned only 1/4 Identity.
The solution is to remove all the above config from the Config repo and to add it instead in each App in \src\main\resources\application.yaml.
Ex for the vets-service, 1 & only 1 property-source is declared using 1 client-id only ${VETS_SVC_APP_IDENTITY_CLIENT_ID} :
spring:
cloud:
azure:
#profile: # spring.cloud.azure.profile
# subscription-id:
# tenant-id: ${SPRING_CLOUD_AZURE_TENANT_ID}
#credential:
#managed-identity-enabled: true
keyvault:
secret:
enabled: true
property-source-enabled: true
# endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
property-sources:
- name: kv-cfg-vets # KV Config for each App Vets-Service
endpoint: ${SPRING_CLOUD_AZURE_KEY_VAULT_ENDPOINT}
credential:
managed-identity-enabled: true
client-id: ${VETS_SVC_APP_IDENTITY_CLIENT_ID}
# client-secret: ${AZURE_CLIENT_SECRET} for SPN not for MI
# profile:
# tenant-id: ${SPRING_CLOUD_AZURE_TENANT_ID}
profiles:
active: mysql
The issue tracker is the preferred channel for bug reports, features requests and submitting pull requests.
For pull requests, editor preferences are available in the editor config for easy use in common text editors. Read more and download plugins at http://editorconfig.org.
https://github.com/ezYakaEagle442/azure-spring-apps-petclinic-mic-srv has been forked from https://github.com/Azure-Samples/spring-petclinic-microservices, itself already forked from https://github.com/spring-petclinic/spring-petclinic-microservices
It is not possible to fork twice a repository using the same user account. However you can duplicate a repository
This repo https://github.com/ezYakaEagle442/aks-java-petclinic-mic-srv has been duplicated from https://github.com/spring-petclinic/spring-petclinic-microservices