PROFESSIONAL-CLOUD-DEVELOPER · Question #361
Your application team is developing an ecommerce application. Your team has developed a new functionality that has a dependency on a third-party service. This third-party service will be deployed in a
The correct answer is C. Use a feature flag to enable the new functionality to users on demand. Gradually enable the new. A feature flag is the safest strategy when a dependency's reliability is uncertain. The new functionality can be deployed to production but kept disabled by default. It can be gradually enabled for a small percentage of users, monitored for issues, and-critically-disabled instant
Question
Your application team is developing an ecommerce application. Your team has developed a new functionality that has a dependency on a third-party service. This third-party service will be deployed in a few days. However, you have been unable to ensure the reliability of this service. You need to choose a deployment strategy for the ecommerce application that will avoid disruption and can be rolled back quickly if issues are discovered. What should you do?
Options
- ADeploy the new functionality by using an A/B deployment strategy.
- BDeploy the new functionality to all users by using a blue/green deployment strategy.
- CUse a feature flag to enable the new functionality to users on demand. Gradually enable the new
- DGradually roll out the new functionality by using a rolling updates deployment strategy. Start with
How the community answered
(22 responses)- A9% (2)
- B5% (1)
- C82% (18)
- D5% (1)
Explanation
A feature flag is the safest strategy when a dependency's reliability is uncertain. The new functionality can be deployed to production but kept disabled by default. It can be gradually enabled for a small percentage of users, monitored for issues, and-critically-disabled instantly without any redeployment if the third-party service fails. This gives the fastest possible rollback (a single configuration change). Option A (A/B deployment) is for comparing two variants, not for controlled risk mitigation around unreliable dependencies. Option B (blue/green) deploys to all users at once and rollback requires switching all traffic back, which is slower. Option D (rolling updates) gradually replaces instances but does not allow instant rollback; you'd need to deploy again to revert.
Topics
Community Discussion
No community discussion yet for this question.