Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Database Engineer practice bank has an additional 769 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
An enterprise runs a critical self-managed SQL Server workload on Compute Engine with synchronous replication across zones in us-central1 and asynchronous replication to us-east1. Business continuity SLAs define these targets:
Before a major product launch, you need to simulate both zonal and regional service disruptions in a staging environment that mirrors production to validate the disaster recovery (DR) workflows and assess compliance.
Which testing and measurement approach should you use?
us-east1 replica to read-write while allowing test clients to continue writing to us-central1, and record only the database promotion duration as the RTO without running snapshot comparisons.Promoting a secondary replica in us-east1 to accept read-write operations transitions its role to a new primary database. However, leaving the original primary in us-central1 accessible to client writes creates an active dual-primary state.
us-central1 and us-east1 produces split-brain data divergence, invalidating the test environment and masking real failover behavior.No existing scenario makes dual active writes valid for a standard active-passive SQL Server availability group; client access must always be halted or fenced before promoting a secondary replica.
During any switchover or failover drill, client write traffic must be stopped or blocked at the network layer before role promotion so that exactly one primary database processes state changes.
Reviewing Cloud Monitoring replication lag and checking backup job completion schedules verifies that data synchronization and snapshot pipelines are operating normally under steady-state conditions. It serves as ongoing health monitoring rather than an active DR simulation.
us-east1 is crash-consistent or free of corruption after an abrupt outage.The goal were simply to perform day-to-day operational monitoring of replication health and backup compliance between scheduled quarterly DR drills.
Because disaster recovery is not part of normal daily operations, relying solely on theoretical estimates creates hidden operational risk. Active switchover or chaos engineering drills are essential to uncover configuration drift and measure true RTA and RPA.
us-central1; during each failover, prevent split-brain writes, compare pre- and post-failure snapshots to verify recovery point actual (RPA), and measure end-to-end recovery time actual (RTA).Comprehensive disaster recovery (DR) simulation involves intentionally injecting realistic infrastructure failures—covering both zonal and regional failure domains—in an isolated, production-equivalent staging environment to validate failover runbooks and measure real-world recovery metrics against defined Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO).
us-central1. Using firewall rules or API restrictions to sever connectivity to us-central1 simulates a complete regional outage and validates cross-region failover to us-east1.Unlike passive monitoring or partial storage tests, this approach exercises the full failure-detection, failover, traffic-rerouting, and data-verification lifecycle for both failure domains without endangering production data.
force-attach operation on a Regional Persistent Disk to move the database to a standby VM in a secondary zone within us-central1 to validate both SLA tiers, recording the disk attachment duration as the end-to-end RTO.force-attach doesThe force-attach operation detaches a Regional Persistent Disk replica from an unresponsive VM in a failed zone and attaches the synchronously replicated copy to a standby VM in another zone within the same region.
us-central1). If an entire region suffers an outage, both zonal replicas become unavailable, so this test fails to validate the regional outage workflow or cross-region failover to us-east1.force-attach in under one minute, but recording only disk attachment time severely underestimates end-to-end RTO.The objective were strictly to validate storage-layer failover latency for a single-zone outage using Regional Persistent Disks, rather than assessing full-stack zonal and regional DR compliance.
Validating single point of failure domains requires separate tests for zonal and regional scopes. Storage failover speed is only one component of recovery time actual (RTA), which must encompass the entire application stack.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated