AWS DevOps Release Strategies 3 — Questions and Answers
Question 1: A company wants to test a new feature with only users in the 'premium' tier before a general rollout. Which release strategy best describes this approach?
- Canary deployment
- Shadow deployment
- Feature flag with targeted rollout (Correct answer)
- A/B testing deployment
Correct answer: Feature flag with targeted rollout
Feature flags with targeted rollout allow specific user segments (like premium users) to access new functionality independently of deployment to all users.
Question 2: In a shadow deployment, traffic is sent to both the old and new service versions. What is the primary goal?
- To let users choose which version they prefer
- To compare the behavior and performance of the new version under real traffic without affecting users (Correct answer)
- To serve 50% of users on each version for statistical comparison
- To validate that the new version's infrastructure can scale under load
Correct answer: To compare the behavior and performance of the new version under real traffic without affecting users
Shadow deployments mirror real production traffic to the new version so its responses and performance can be compared against the current version without impacting users.
Question 3: When using Amazon Route 53 weighted routing for a blue/green deployment, what value should the blue environment's weight be set to after a successful cutover?
- 0 (Correct answer)
- 50
- 100
- It should be deleted immediately
Correct answer: 0
Setting the blue environment's weight to 0 removes it from rotation while keeping the DNS record intact, enabling a fast rollback by simply increasing its weight again.
Question 4: A development team is implementing A/B testing in their release pipeline. What is the fundamental difference between A/B testing and a standard canary deployment?
- A/B testing measures user behavior metrics to make product decisions; canary deployments focus on technical health metrics to ensure stability (Correct answer)
- A/B testing always uses 50/50 traffic splits; canary uses 5/95 splits
- A/B testing requires AWS CloudWatch; canary deployments require AWS X-Ray
- Canary deployments are only for Lambda; A/B testing is only for EC2
Correct answer: A/B testing measures user behavior metrics to make product decisions; canary deployments focus on technical health metrics to ensure stability
A/B testing compares user experience and conversion metrics between variants to inform product decisions, while canary deployments monitor system health to validate safe progressive rollout.
Question 5: Which AWS CodeDeploy deployment configuration would shift 10% of traffic immediately and then shift the remaining 90% after 30 minutes?
- CodeDeployDefault.LambdaLinear10PercentEvery10Minutes
- CodeDeployDefault.LambdaCanary10Percent30Minutes (Correct answer)
- CodeDeployDefault.LambdaAllAtOnce
- CodeDeployDefault.HalfAtATime
Correct answer: CodeDeployDefault.LambdaCanary10Percent30Minutes
The LambdaCanary10Percent30Minutes configuration shifts 10% of traffic to the new version immediately and waits 30 minutes before shifting the remaining 90%.
Question 6: What is a key advantage of using AWS CloudFormation StackSets for multi-region deployments as part of a release strategy?
- StackSets automatically perform canary rollouts across regions in sequence
- StackSets allow a single CloudFormation template to be deployed across multiple accounts and regions with centralized management (Correct answer)
- StackSets provide built-in rollback to the previous version when a CloudWatch alarm fires
- StackSets integrate with CodeDeploy to shift traffic between regions
Correct answer: StackSets allow a single CloudFormation template to be deployed across multiple accounts and regions with centralized management
CloudFormation StackSets enable deploying a single template to multiple AWS accounts and regions simultaneously, simplifying coordinated multi-region release operations.
Question 7: In an ECS blue/green deployment managed by AWS CodeDeploy, what resource handles the traffic shift between the old and new task set?
- An Amazon Route 53 weighted record
- An Application Load Balancer listener rule with two target groups (Correct answer)
- An AWS Global Accelerator endpoint group
- An Amazon CloudFront distribution with origin failover
Correct answer: An Application Load Balancer listener rule with two target groups
CodeDeploy controls the ALB listener to shift traffic between the production target group (blue) and replacement target group (green) during an ECS blue/green deployment.
A company wants to test a new feature with only users in the 'premium' tier before a general rollout.
Which release strategy best describes this approach?