A company updated the AWS CloudFormation template for a critical business application. The stack update process failed due to an error in the updated template, and CloudFormation automatically began the stack rollback process. Later, a DevOps engineer found the application was still unavailable, and that the stack was in the UPDATE_ROLLBACK_FAILED state.
Which combination of actions will allow the stack rollback to complete successfully? (Choose two.)

A. Attach the AWSCloudFormationFullAccess IAM policy to the CloudFormation role.
B. Automatically heal the stack resources using CloudFormation drift detection.
C. Issue a ContinueUpdateRollback command from the CloudFormation console or AWS CLI.
D. Manually adjust the resources to match the expectations of the stack.
E. Update the existing CloudFormation stack using the original template.