SOA-C02 Question 153
Single answerA company wants to automate the deployment of a web application across multiple environments (e.g., development, staging, and production) in a consistent and repeatable way. They also want to centralize configuration management and avoid manually updating environment-specific parameters. Which approach will best achieve these requirements?
- A
Use AWS Systems Manager Parameter Store to store environment-specific configurations and AWS CloudFormation templates to define and deploy the application stack.
- B
Manually update application configurations in each environment and deploy the application using AWS Elastic Beanstalk.
- C
Use AWS Lambda functions to manually trigger deployment scripts for each environment and manage configurations in Amazon S3.
- D
Deploy the application using the AWS Management Console to set up resources in each environment and manually configure parameters.
Show answer and explanation
Correct answer: A
Explanation
The combination of AWS Systems Manager Parameter Store and AWS CloudFormation fulfills the requirements for automating deployment, centralizing configuration management, and ensuring consistency across multiple environments. Parameter Store allows you to manage environment-specific settings securely, while CloudFormation provides a way to define infrastructure as code for repeatable deployments.
- A. Correct.
Correct. AWS Systems Manager Parameter Store can centralize and securely manage environment-specific parameters, while CloudFormation templates ensure consistent and repeatable deployment of resources across environments.
- B. Incorrect.
Incorrect. Manually updating configurations is error-prone and not aligned with the requirement for automation and consistency.
- C. Incorrect.
Incorrect. While AWS Lambda can trigger deployment scripts, this approach does not centralize configuration management or ensure consistency across environments.
- D. Incorrect.
Incorrect. Manually deploying resources and configuring parameters is time-consuming, prone to human error, and does not meet the requirement for automation.