Question

Difficulty: HardDeployment Strategy Design

A company uses AWS CloudFormation StackSets to manage infrastructure deployments across 50 production accounts and 1 staging account in an AWS Organization. The Solutions Architect needs to design a deployment strategy that updates the stacks in these accounts. The strategy must meet the following requirements:

- The stack update must be applied to the staging account and verified successfully before any production accounts are updated.
- In production, updates must be deployed to the us-east-1 Region first, and must only proceed to the eu-west-1 Region if the deployment in us-east-1 is completely successful.
- Within each Region, the deployment must update a maximum of 10 production accounts concurrently.
- If deployments fail in more than 2 production accounts in a single Region, the deployment operation for that Region must immediately stop, and the stack set must roll back.

Which of the following actions should the Solutions Architect take to meet these requirements? (Select TWO.)

  1. Deploy the CloudFormation template to the staging account first. Once verified, execute the StackSet update operation targeting the production accounts with the RegionConcurrencyType parameter set to SEQUENTIAL and the RegionOrder parameter set to ['us-east-1', 'eu-west-1'].Answer
  2. Configure the StackSet update operation preferences for the production accounts with the MaxConcurrentCount parameter set to 10 and the FailureToleranceCount parameter set to 2.Answer
  3. C
    Configure a single StackSet targeting an organizational unit (OU) that contains both the staging and production accounts, and set the MaxConcurrentPercentage parameter to 2% to ensure staging is updated first.
  4. D
    Configure the production StackSet update operation preferences with the RegionConcurrencyType parameter set to PARALLEL and set the RegionOrder parameter to ['us-east-1', 'eu-west-1'] to allow concurrent deployment across Regions.
  5. E
    Configure the production StackSet update operation preferences with the FailureToleranceCount parameter set to 0 to trigger immediate rollback upon any single account deployment failure.

Answer

Deploy the CloudFormation template to the staging account first, then execute the StackSet update targeting production accounts with the RegionConcurrencyType parameter set to SEQUENTIAL and RegionOrder set to ['us-east-1', 'eu-west-1']. Configure the StackSet update operation preferences with the MaxConcurrentCount parameter set to 10 and the FailureToleranceCount parameter set to 2.
To satisfy the deployment requirements, the staging account must be updated and verified independently before production. For the production accounts, AWS CloudFormation StackSets operational preferences must be configured to control the order, concurrency, and failure threshold. Setting the RegionConcurrencyType to SEQUENTIAL with a RegionOrder of ['us-east-1', 'eu-west-1'] ensures that us-east-1 is fully deployed before eu-west-1 is updated. Setting the MaxConcurrentCount to 10 limits the active accounts being updated per Region to 10. Setting the FailureToleranceCount to 2 allows up to 2 failures per Region; if a 3rd failure occurs, the operation immediately stops and rolls back.

Step-by-Step Solution

1
Separate the deployment stages to ensure validation.
The staging account is updated first using a direct stack update or separate StackSet operation.
This guarantees staging validation is completed before production accounts are modified.
2
Enforce regional sequence for the production StackSet update.
The RegionConcurrencyType parameter is set to SEQUENTIAL, and the RegionOrder parameter is set to ['us-east-1', 'eu-west-1'].
This ensures the deployment in us-east-1 completes successfully before proceeding to eu-west-1.
3
Set the parallel execution limits and rollback triggers for production accounts.
The MaxConcurrentCount parameter is set to 10, and the FailureToleranceCount parameter is set to 2.
This limits concurrent deployments to 10 accounts and allows up to 2 failures, stopping and rolling back the operation if a 3rd failure occurs (more than 2 failures).

Key Concept

AWS CloudFormation StackSets concurrency, regional routing, and failure tolerance controls.
Estimated Time:3m 0s
Rate this question