In this lab we'll take a look at an application called Globo Tickets. It's an application that consists of 3 services working together. First of all there is a front end from their URLs serving a web shop as a website. This front end from their URLs uses the catalog from their URLs to retrieve the items & prices to populate the webshop. After that whenever an order is placed this is sent to an ordering from their URLs which will handle further processing of this order.
To get a good feeling of the application you can run the application locally before we actually deploy it in the cloud. We can do this by using Docker Compose.
There is a Docker Compose file (docker-compose.yml) in the src/globo-tickets-basic directory that you could use to run the app locally.
Installing Docker Compose:
- If you have Docker Desktop installed you already have Docker Compose ready to use. If not you can install the compose plugin next to Docker here: https://docs.docker.com/compose/install/
Tip for running Docker Compose:
In Lab 1 - My first container app we've seen all the requirements we need to create an Azure Container App Environment. Before we can deploy our application we need to create a:
- Azure Log Analytics Workspace
- Azure Container App Environment
- (optional) Application Insights
Since we're not cavemen/women we will create these resources using Infrastructure as Code. You are free to choose your own language if you dare but in this lab we'll describe how to do this through Azure CLI or using Bicep.
If you are up for the challenge of doing this through Pulumi or Terraform be our guest! we would love ❤️ to see what you did so we can integrate it into these labs to make it better. PRs welcome ;)
To create the required resources through CLI you can take a look at the following az cli commands:
az monitor log-analytics workspace createaz containerapp env createaz monitor app-insights component create
Add them in a script file so you can run this automation more often (for example from a pipeline).
To create the app environment + additional resources you could use the following piece of bicep to create a re-usable module.
note that we create an output parameter with the Id of the Azure Container App Environment to be used later in other modules.
param environmentName string = 'globotickets'
param logAnalyticsWorkspaceName string = 'globotickets'
param appInsightsName string = 'globotickets'
param location string = resourceGroup().location
resource logAnalyticsWorkspace 'Microsoft.OperationalInsights/workspaces@2020-03-01-preview' = {
name: logAnalyticsWorkspaceName
location: location
properties: any({
retentionInDays: 30
features: {
searchVersion: 1
}
sku: {
name: 'PerGB2018'
}
})
}
resource appInsights 'Microsoft.Insights/components@2020-02-02' = {
name: appInsightsName
location: location
kind: 'web'
properties: {
Application_Type: 'web'
WorkspaceResourceId:logAnalyticsWorkspace.id
}
}
resource environment 'Microsoft.App/managedEnvironments@2022-03-01' = {
name: environmentName
location: location
properties: {
daprAIInstrumentationKey:appInsights.properties.InstrumentationKey
appLogsConfiguration: {
destination: 'log-analytics'
logAnalyticsConfiguration: {
customerId: logAnalyticsWorkspace.properties.customerId
sharedKey: logAnalyticsWorkspace.listKeys().primarySharedKey
}
}
//vnet config
}
}
output environmentId string = environment.id
In this example we're not adding a connection to an existing vnet. You could do this yourself by adding the required properties to do so in the line that is commented out that states
//vnet config
Now that we have an empty environment we can deploy the application to it. It consists of 3 container images. These containers have different requirments.
- Frontend Link to Image (Has to have external ingress)
- Catalog Link to Image (Has to have internal ingress)
- Ordering Link to Image (Has to have internal ingress)
All examples here container a sample through the CLI or Bicep. choose either one of them which you prefer.
Creating Container apps in a Container App Environment can be done through this command
Since all container apps use the same resource we can create a module deploying the container app. A sample module is listed below that you have to provision 3x so each part of the application is deployed.
param location string = resourceGroup().location
param environmentId string
param containerAppName string
param containerRegistry string
param containerRegistryUsername string
param registryPassword string
param ingressIsExternal bool
param image string
param environmentVariables array
param secrets array = []
resource containerApp 'Microsoft.App/containerApps@2022-03-01' = {
name: containerAppName
location: location
properties: {
managedEnvironmentId: environmentId
configuration: {
activeRevisionsMode: 'Single'
secrets: secrets
registries:[
{
server: containerRegistry
username: containerRegistryUsername
passwordSecretRef: registryPassword
}
]
ingress: {
external: ingressIsExternal
targetPort: 80
}
dapr: {
enabled: true
appPort: 80
appId: containerAppName
}
}
template: {
containers: [
{
image: image
name: containerAppName
env: environmentVariables
}
]
scale: {
minReplicas: 1
maxReplicas: 1
}
}
}
}
output url string = 'https://${containerApp.properties.configuration.ingress.fqdn}'
If you go to the Azure portal now you should see your Container app environment with the running apps in it. Navigate to the resource group where you deployed the container app environment and select the environment. You should now see all 3 apps running in that environment.
If you select the frontend application you can view the details of the app. On the top right you'll find the public URL of the application.
If you browse to the url of the frontend application you should be seeing a running application. Buy some tickets to see if everything works!
Try accessing the catalog & ordering service from their URLs. These are set to internal so the Application URL should not be accessible from your machine.
