[Developer] Deployment Services

Question 1

What is AWS CodeDeploy?

Answer 1

A service that automates code deployments to a variety of compute services (EC2, Lambda, ECS, and Serverless).

It is infrastructure-agnostic.

Question 2

What is the key configuration file used by CodeDeploy for EC2/on-premises deployments?

Answer 2

The appspec.yml file.

Question 3

What is the purpose of the appspec.yml file?

Answer 3

It specifies the source files to copy and the lifecycle event hooks (scripts) to run during deployment.

Question 4

What are the three main deployment configurations for an EC2/on-premises deployment?

Answer 4

1. In-place
2. Blue/Green
3. Canary (for Lambda/ECS/EC2)

Question 5

Which CodeDeploy deployment type replaces the old instances with a fresh set of new instances?

Answer 5

Blue/Green deployment.

Question 6

What is the main benefit of a Blue/Green deployment over an In-Place deployment?

Answer 6

Zero downtime and instant rollback capability.

The entire environment is tested before traffic is shifted.

Question 7

In an In-Place deployment, where is the application stopped, updated, and re-started?

Answer 7

Directly on the existing instances.

Question 8

Where must the CodeDeploy Agent be installed for EC2 and on-premises deployments?

Answer 8

Directly on the EC2 instance or on-premises server.

Question 9

What is the recommended deployment strategy for a Lambda function in CodeDeploy?

Answer 9

Canary (or Linear)

incrementally shifts traffic to the new version, using CloudWatch Alarms to monitor and auto-rollback on failure.

Question 10

What is AWS Elastic Beanstalk (EB)?

Answer 10

A high-level Platform as a Service (PaaS) that handles the complexity of deploying, managing, and scaling web applications (including provisioning EC2, load balancers, etc.).

Question 11

What is the key benefit of Elastic Beanstalk for a developer?

Answer 11

It allows the developer to focus only on writing code; Beanstalk handles all the underlying infrastructure management.

Question 12

What kind of files do you provide to Beanstalk?

Answer 12

Only your application code/package (e.g., a .zip or .jar file).

Question 13

Which tool does Beanstalk use under the hood to provision and update the underlying resources?

Answer 13

AWS CloudFormation.

Question 14

How does a developer configure custom infrastructure settings (e.g., install specific packages or set environment variables) in a Beanstalk environment?

Answer 14

By adding .ebextensions configuration files to the application source code bundle.

Question 15

If you need to perform a database migration during a Beanstalk deployment, which deployment type should you use to ensure safety and zero downtime?

Answer 15

Blue/Green (or possibly Immutable).

Question 16

What is the difference between an EB Environment and an EB Application?

Answer 16

An Application is a logical collection of components (templates, versions). An Environment is a specific, running instance of an Application Version (e.g., production or staging).

Question 17

What deployment option does EB use that involves launching new instances with the new code and then terminating the old ones?

Answer 17

Immutable deployment (provides a safe rollback and zero downtime).

Question 18

What is AWS CodePipeline?

Answer 18

A Continuous Delivery (CD) service that automates the release process for application and infrastructure updates, following a defined workflow.

Question 19

What are the 2 core components of a CodePipeline workflow?

Answer 19

1. Stages (e.g., Source, Build, Deploy)
2. Actions (the tasks performed within a stage).

Question 20

What is the typical flow of data between stages in CodePipeline?

Answer 20

Artifacts.

The output artifact of one stage is the input artifact of the next stage.

Question 21

Where does CodePipeline store the artifacts (source code, build files) that pass between stages?

Answer 21

An S3 bucket

(created and managed by the pipeline itself).

Question 22

In CodePipeline

What is the purpose of the Source Stage?

Answer 22

To pull the latest code from a source repository (e.g., CodeCommit, GitHub, S3) and place it into the S3 artifact store.

Question 23

In CodePipeline

What is the primary AWS service used in the Build Stage?

Answer 23

AWS CodeBuild.

Question 24

In CodePipeline

What is the primary AWS service used in the Deploy Stage when deploying to EC2/on-premises?

Answer 24

AWS CodeDeploy.

Question 25

How can you introduce a manual approval step into a CodePipeline pipeline?

Answer 25

By adding an **Approval Action** as a step in any stage of the pipeline.

Question 26

How does CodePipeline integrate with CloudFormation to deploy infrastructure?

Answer 26

Using the **CloudFormation Action**, which can create, update, or delete a CloudFormation stack.

Question 27

What is the simplest way to roll back a failing deployment in CodePipeline?

Answer 27

Re-run the pipeline using a previous, successful revision from the Source Stage.

Question 28

How does CodePipeline handle a failure in any stage?

Answer 28

The pipeline **stops execution** at the point of failure, and the deployed resources are left in their current state (no automatic rollback).

Question 29

How can a developer use Lambda for testing or automation within a CodePipeline stage?

Answer 29

Use a CodePipeline action that invokes a Lambda function to perform custom tasks (e.g., security checks, custom notifications).

Question 30

How does CodePipeline roll back existing resources after a failure part of the way through a deployment?

Answer 30

There is **no automatic rollback**, but you can **configure lifecycle hooks** in the `appspec.yml` file for cleanup.

Question 31

How do you configure a CodePipeline in Account A to deploy resources into Account B?

Answer 31

Configure an **IAM Cross-Account Role** in Account B that has the necessary deployment permissions and trusts the CodePipeline Service Role in Account A.

Question 32

When should a developer choose AWS Elastic Beanstalk over AWS Amplify?

Answer 32

Beanstalk is for **traditional, server-based applications** (e.g., Tomcat, Python, Docker) where the developer needs some control over the EC2/OS level, but still wants managed infrastructure.

Question 33

What `appspec.yaml` lifecycle hook is used for **graceful stopping** of an application?

Answer 33

`ApplicationStop`

Question 34

A developer is implementing CI/CD using CodePipeline. The source stage uses CodeCommit. How should the pipeline be automatically triggered when code is pushed to the repository?

Answer 34

Use an Amazon CloudWatch Event Rule (or EventBridge) listening for CodeCommit events.

Question 35

What exactly is the benefit of the deployment strategy of "rolling _with an additional batch_"?

Answer 35

It guarantees that the Auto Scaling Group capacity is never reduced during the deployment

Question 36

Which component is responsible for rerouting traffic from the 'Blue' (old) environment to the 'Green' (new) environment?

Answer 36

CodeDeploy uses the **load balancer (ALB/NLB)** to swap the target group associated with the production listener from the old (Blue) environment to the new (Green) environment.

Question 37

Which deployment type in CodeDeploy is required for AWS Lambda functions?

Answer 37

Blue/Green

Question 38

What Cloudformation intrinsic is used to get the **canonical name or id** of a resource?

Answer 38

`!Ref`

Question 39

What Cloudformation intrinsic is used to get **metadata** about a resource?

Answer 39

`!GetAtt`

Question 40

What Cloudformation Intrinsic is used to get data **from another stack**?

Answer 40

`!ImportValue`