Why containers on Azure matter - Why It Works

Introduction

Containers help you package your app and everything it needs to run in one box. Azure lets you run these containers easily in the cloud, so your app works the same everywhere and can grow when many people use it.

When you want to move your app to the cloud without changing how it works on your computer.

When you need to run many copies of your app to handle lots of users at the same time.

When you want to update your app often without stopping it for a long time.

When you want to save money by using only the cloud resources your app needs at the moment.

When you want to keep your app safe and separate from other apps running on the same cloud.

Commands

This command creates a container registry in Azure where you can store your container images safely.

Terminal

az acr create --resource-group example-group --name exampleRegistry --sku Basic

Expected Output

{

  "adminUserEnabled": false,

  "creationDate": "2024-06-01T12:00:00Z",

  "id": "/subscriptions/xxxx/resourceGroups/example-group/providers/Microsoft.ContainerRegistry/registries/exampleRegistry",

  "location": "eastus",

  "loginServer": "exampleRegistry.azurecr.io",

  "name": "exampleRegistry",

  "resourceGroup": "example-group",

  "sku": {

    "name": "Basic",

    "tier": "Basic"

  },

  "status": "Succeeded",

  "type": "Microsoft.ContainerRegistry/registries"

}

→

--resource-group - Specifies the Azure group where the registry will be created

→

--name - Sets the unique name for your container registry

→

--sku - Chooses the pricing tier for the registry

This command logs you into the Azure container registry so you can push and pull container images.

Terminal

az acr login --name exampleRegistry

Expected Output

Login Succeeded

→

--name - Specifies which container registry to log into

This command builds your container image from your app code and tags it with the registry address and version.

Terminal

docker build -t exampleRegistry.azurecr.io/my-app:v1 .

Expected Output

Sending build context to Docker daemon  10.24kB

Step 1/5 : FROM node:18-alpine

 ---> 123abc456def

Step 2/5 : WORKDIR /app

 ---> Using cache

 ---> 789ghi012jkl

Step 3/5 : COPY package*.json ./

 ---> Using cache

 ---> mno345pqr678

Step 4/5 : RUN npm install

 ---> Running in abc123def456

Removing intermediate container abc123def456

 ---> stu901vwx234

Step 5/5 : COPY . .

 ---> Using cache

 ---> yz567abc890

Successfully built yz567abc890

Successfully tagged exampleRegistry.azurecr.io/my-app:v1

This command uploads your container image to the Azure container registry so it can be used in the cloud.

Terminal

docker push exampleRegistry.azurecr.io/my-app:v1

Expected Output

The push refers to repository [exampleRegistry.azurecr.io/my-app]

123abc456def: Pushed 

789ghi012jkl: Pushed 

mno345pqr678: Pushed 

stu901vwx234: Pushed 

yz567abc890: Pushed 

v1: digest: sha256:abcdef1234567890 size: 1578

This command creates a container instance in Azure to run your app with 1 CPU, 1.5 GB memory, and makes it reachable on the internet with a DNS name.

Terminal

az container create --resource-group example-group --name myAppContainer --image exampleRegistry.azurecr.io/my-app:v1 --cpu 1 --memory 1.5 --dns-name-label myappdnslabel --ports 80

Expected Output

{

  "id": "/subscriptions/xxxx/resourceGroups/example-group/providers/Microsoft.ContainerInstance/containerGroups/myAppContainer",

  "name": "myAppContainer",

  "location": "eastus",

  "properties": {

    "provisioningState": "Succeeded",

    "containers": [

      {

        "name": "myAppContainer",

        "properties": {

          "image": "exampleRegistry.azurecr.io/my-app:v1",

          "resources": {

            "requests": {

              "cpu": 1.0,

              "memoryInGb": 1.5

            }

          },

          "ports": [

            {

              "port": 80

            }

          ]

        }

      }

    ],

    "ipAddress": {

      "dnsNameLabel": "myappdnslabel",

      "ports": [

        {

          "port": 80,

          "protocol": "TCP"

        }

      ],

      "type": "Public"

    }

  },

  "type": "Microsoft.ContainerInstance/containerGroups"

}

→

--cpu - Sets how many CPUs the container can use

→

--memory - Sets how much memory the container can use

→

--dns-name-label - Creates a public DNS name to access the container

Key Concept

If you remember nothing else from this pattern, remember: containers let you package your app once and run it anywhere in Azure with the same behavior and easy scaling.

Common Mistakes

Trying to push a container image without logging into the Azure container registry first

The push will fail because Docker does not have permission to upload the image

Always run 'az acr login --name exampleRegistry' before pushing images

Not tagging the container image with the full registry login server address before pushing

Docker will try to push to Docker Hub or another default registry and fail

Tag images with 'exampleRegistry.azurecr.io/my-app:tag' before pushing

Creating a container instance without specifying ports or DNS name

Your app will run but you won't be able to access it from the internet

Use '--ports' and '--dns-name-label' flags to expose your app publicly

Summary

Create an Azure container registry to store your container images securely.

Build and tag your container image with the registry address.

Log in to the registry and push your container image.

Create a container instance in Azure to run your app with specified resources and public access.

Practice

1. Why are containers useful when running applications on Azure?

easy

A. They only work on Windows servers.

B. They require more hardware than traditional apps.

C. They make apps easy to move and run anywhere.

D. They need manual setup for each server.

Solution

1. Step 1: Understand container portability

   Containers package apps with everything needed, so they run the same anywhere.

2. Step 2: Compare with other options

   Unlike manual setups or OS-specific apps, containers simplify moving and running apps.

3. Final Answer:

   They make apps easy to move and run anywhere. -> Option C

4. Quick Check:

   Containers = portability [OK]

Hint: Containers bundle apps for easy movement and running [OK]

Common Mistakes:

• Thinking containers need more hardware
• Believing containers only run on Windows
• Assuming manual setup is always required

2. Which Azure service is designed specifically to run containers easily?

easy

A. Azure Virtual Machines

B. Azure Kubernetes Service

C. Azure Blob Storage

D. Azure SQL Database

Solution

1. Step 1: Identify container-focused services

   Azure Kubernetes Service (AKS) is built to manage and run containers at scale.

2. Step 2: Eliminate unrelated services

   Virtual Machines run full OS, Blob Storage stores files, SQL Database manages data, none focus on containers.

3. Final Answer:

   Azure Kubernetes Service -> Option B

4. Quick Check:

   AKS = container management [OK]

Hint: AKS is Azure's container orchestration service [OK]

Common Mistakes:

• Confusing VMs with container services
• Choosing storage or database services
• Not knowing AKS purpose

3. What happens when you deploy a containerized app on Azure Container Instances (ACI)?

medium

A. Azure automatically provisions compute resources and runs the container.

B. You must manually configure virtual machines before deployment.

C. The app runs only on your local machine, not in the cloud.

D. Azure converts the container into a virtual machine image.

Solution

1. Step 1: Understand Azure Container Instances behavior

   ACI lets you run containers without managing servers; Azure handles resources automatically.

2. Step 2: Compare other options

   Manual VM setup or local-only running is not how ACI works; it does not convert containers to VM images.

3. Final Answer:

   Azure automatically provisions compute resources and runs the container. -> Option A

4. Quick Check:

   ACI = serverless container run [OK]

Hint: ACI runs containers without manual VM setup [OK]

Common Mistakes:

• Thinking manual VM setup is needed
• Believing containers run only locally
• Confusing containers with VM images

4. You tried to deploy a container on Azure but it failed. Which common mistake might cause this?

medium

A. Using too much memory in the container.

B. Deploying without an internet connection.

C. Running the container on a Windows machine.

D. Not specifying the container image name correctly.

Solution

1. Step 1: Identify common deployment errors

   Incorrect container image names cause deployment failures because Azure cannot find the image.

2. Step 2: Evaluate other options

   Memory limits cause runtime issues, not deployment failure; Windows machines can run containers; internet is needed but usually checked beforehand.

3. Final Answer:

   Not specifying the container image name correctly. -> Option D

4. Quick Check:

   Wrong image name = deployment fail [OK]

Hint: Check container image name spelling first [OK]

Common Mistakes:

• Ignoring image name typos
• Confusing runtime errors with deployment errors
• Assuming OS blocks deployment

5. How do containers on Azure help save money and time when scaling an app?

hard

A. They use resources efficiently and start quickly without full OS boot.

B. They require buying extra hardware for each container.

C. They force manual updates on every server.

D. They run only one app per server, increasing costs.

Solution

1. Step 1: Understand container resource use

   Containers share the OS kernel, so they use less memory and CPU than full virtual machines.

2. Step 2: Understand startup and scaling benefits

   Containers start fast without booting an OS, enabling quick scaling and saving time and money.

3. Final Answer:

   They use resources efficiently and start quickly without full OS boot. -> Option A

4. Quick Check:

   Containers = efficient, fast scaling [OK]

Hint: Containers share OS, start fast, save costs [OK]

Common Mistakes:

• Thinking containers need extra hardware
• Believing manual updates are required
• Assuming one app per server