PCA Question 81

Community Discussion

Written by other learners. Comments can disagree with the marked answer — check the official documentation when in doubt.

• Chose C1

  Regional MIG Resilience: A regional managed instance group is designed to survive zonal outages by automatically redistributing or recreating instances in healthy zones. Shutting off all VMs in a zone is a standard "chaos engineering" practice to validate that your application remains available and that your autoscaling/self-healing configurations work as expected under stress. Google-Recommended Practice: Google explicitly provides scripts and metadata methods to simulate zonal failures for regional MIGs to test if they are correctly overprovisioned and capable of surviving such events. .

• 1

  Targeting the Architecture: Since your application uses a regional MIG, testing its ability to handle a full zone being offline is the most comprehensive way to validate the "resilience" of the authentication layer. not D: While this tests the database's High Availability (HA) failover, a regional managed instance group is primarily an infrastructure-level tool. Testing the VMs themselves (Option C) provides a broader validation of the entire application stack's ability to reroute traffic and maintain service during a disaster. In a true resilience test, you want to see how the application handles the combined loss of compute and potential database reconnects.

• 1

  Resilience is a broader concept. It is the ability of a system to not only survive a failure but also to adapt to and recover quickly from adverse conditions or disruptions. It's about how the system handles unforeseen failures and operates even in a degraded state. Key Question: Can the system recover from failure? How it's achieved: Through strategies like self-healing, graceful degradation, and a robust disaster recovery plan. A resilient system may experience a temporary dip in performance, but it can quickly bounce back to an effective operational state.