PROFESSIONAL-CLOUD-DEVOPS-ENGINEER Question #171: Real Exam Question with Answer & Explanation

Question

You support a website with a global audience. The website has a frontend web service and a backend database service that runs on different clusters. All clusters are scaled to handle at least ? of the total user traffic. You use 4 different regions in Google Cloud Platform and Cloud Load Balancing to direct traffic to a region closer to the user. You are applying a critical security patch to the backend database. You successfully patch the database in the first 2 regions, but you make a configuration error while patching Region 3. The unsuccessful patching causes 50% of user requests to Region 3 to time out. You want to mitigate the impact of unsuccessful patching on users. What should you do?

Options

• AAdd more capacity to the frontend of Region 3.
• BRevert the Region 3 backend database and run it without the patch.
• CDrain the requests to Region 3 and redirect new requests to other regions.
• DBack up the database in the backend of Region 3 and restart the database.

Explanation

A is incorrect because adding more Frontend servers will not fix the backend configuration B is incorrect because running a backend without the critical security patch is a big risk. C is correct because the remaining 3 regions can handle the total traffic load, which gives you time to fix the configuration error in Region 3 and then apply the patch. D is incorrect because backing up the database will not help at this point, and the additional load on the server could make the problem worse. https://cloud.google.com/load-balancing/docs/enabling-connection-draining https://cloud.google.com/compute/docs/tutorials/robustsystems#distribute https://landing.google.com/sre/sre-book/chapters/addressing-cascading-failures/

No community discussion yet for this question.