AZ-400 · Question #466
AZ-400 Question #466: Real Exam Question with Answer & Explanation
The correct answer is A: a single web app and two deployment slots. Explanation Option A is correct because Azure App Service deployment slots allow you to deploy a new version to a staging slot while keeping the current production version live, then perform a slot swap that instantly redirects traffic to the new version with near-zero downtime.
Question
You have an Azure web app that is deployed by using Azure Pipelines. You need to ensure that when a new version of the app is deployed to production, you can roll back to the previous version. The solution must meet the following requirements: - Minimize downtime during the deployment. - Minimize the time it takes for the rollback. What should you use?
Options
- Aa single web app and two deployment slots
- Ba single web app and two deployment pipelines
- Ctwo web apps and an Azure Standard Load Balancer
- Dtwo web apps and an Azure Traffic Manager instance
Explanation
Explanation
Option A is correct because Azure App Service deployment slots allow you to deploy a new version to a staging slot while keeping the current production version live, then perform a slot swap that instantly redirects traffic to the new version with near-zero downtime. If issues arise, rolling back is as simple as swapping back to the previous slot - a process that takes only seconds, satisfying both the downtime and rollback speed requirements.
Option B is wrong because having two pipelines does nothing to manage traffic routing or enable fast rollbacks - you'd still have downtime during redeployment and no instant rollback mechanism.
Option C is wrong because Azure Standard Load Balancer is designed for infrastructure-level (IaaS) load balancing, not for PaaS web app deployments, and switching traffic between two separate web apps would be slower and more complex than a slot swap.
Option D is wrong because while Azure Traffic Manager can route traffic between two web apps, it relies on DNS propagation for failover, which introduces latency (minutes, not seconds) - making rollback slower than deployment slots.
💡 Memory Tip: Think of deployment slots as a "staging runway" - the plane (new version) lands safely on the staging strip first, and you only redirect it to the main terminal (production) when ready. If there's a problem, you just redirect traffic back instantly!
Topics
Community Discussion
No community discussion yet for this question.