AZ-400 Question 164
Select 2An e-commerce company runs a mission-critical Azure Cloud Service that must remain accessible 24/7. They plan to release a new version of the application and want to minimize downtime during the deployment. They also want to ensure the new version is thoroughly tested in an isolated environment before going live. Which two approaches will help them achieve minimal downtime?
- A
Place the Cloud Service in the Offline state while deploying the updated version, preventing any user access during the process.
- B
Create a staging slot for the Cloud Service, test the new build there, and then perform a VIP swap to move the tested version into production.
- C
Use an Azure Load Balancer in combination with a rolling deployment strategy, updating one instance at a time while the others remain active.
- D
Deploy the new version directly to the production slot, relying on application self-healing to handle any service disruption during the deployment.
- E
Increase the CPU and memory capacity in production to handle the workload spike during a simultaneous upgrade.
Show answer and explanation
Correct answers: B, C
Explanation
Using a staging slot (or staging environment) and then performing a VIP swap provides near-zero downtime deployments by loading a tested version into production almost instantly. Additionally, implementing a rolling deployment strategy with load balancing allows you to update individual instances behind the load balancer while others continue serving user traffic. According to Microsoft� Azure documentation, these two methods are common best practices for minimizing downtime in mission-critical environments.
- A. Incorrect.
Option 1 is incorrect because placing the Cloud Service offline prevents any user access during deployment, causing downtime. This does not align with the goal of minimal or zero downtime.
- B. Correct.
Option 2 is correct because using a staging slot and performing a VIP swap allows you to fully test the new version in an isolated environment before seamlessly swapping it into production, effectively minimizing downtime.
- C. Correct.
Option 3 is correct because using a load balancing solution and a rolling upgrade approach ensures only a portion of instances are updated at any one time, keeping the remaining instances available to serve traffic and reduce downtime risk.
- D. Incorrect.
Option 4 is incorrect because deploying directly to production without an isolated environment can lead to unavoidable downtime or service disruptions if issues arise during the deployment process.
- E. Incorrect.
Option 5 is incorrect because simply increasing capacity does not address the need for isolated testing and a controlled swap. While it may help performance, it doesn't guarantee minimal downtime during the deployment.