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.
A retail company operates an e-commerce platform backed by Cloud SQL for PostgreSQL with separate staging and production instances. Traffic metrics show peak user activity Monday through Saturday, with minimal load on Sundays between 01:00 and 05:00 UTC.
The database team must configure maintenance policies to meet these requirements:
Which maintenance configuration should the team apply to both instances?
This option assigns the Week 2 (production) timing track to the staging environment and the Week 1 (preview) timing track to the production environment, while placing the production maintenance window on Wednesday at 01:00 UTC.
Week 1 applies updates 7 to 14 days after notification and Week 2 applies them 15 to 21 days after notification, this setup updates the live production database roughly a week before the staging database. This defeats the purpose of using staging to validate updates.preview/canary schedule while keeping staging on an older version.This configuration would only make sense in an unusual scenario where the production workload's lowest traffic occurred on Wednesday mornings and the team intentionally wanted production to act as an early canary (which violates standard reliability practices). For standard staging-to-production pipelines, staging must always precede production on an earlier track.
Any and Week 5 Tracks DoIn Cloud SQL, the Any maintenance timing option allows updates to be applied at any time, typically within the first week of a rollout, without enforcing a structured 7-to-14-day waiting period after notification. The Week 5 track represents the most conservative rollout schedule, deferring maintenance updates until 35 to 42 days after the maintenance notification is sent.
Any means the staging instance typically updates within the first week rather than guaranteeing the 7 to 14 days lead time requested by the team after notification.Week 5 pushes production maintenance to 35 to 42 days after notification. Because this exceeds 30 days, it violates the explicit constraint against delaying new features for more than a month.Any unpredictability: Instances with Any receive updates early and potentially at any hour if no specific day and hour are set, leaving little time to prepare test harnesses after an announcement.Week 5 delay: Week 5 delays both new database features and routine non-emergency improvements for five to six weeks.This configuration would be appropriate if the team wanted staging updated as early as possible without needing a 7-to-14-day preparation window, and if production workloads had strict compliance or change-freeze policies requiring over a month of soak time before accepting any new features.
Week 1 Tracks and Maintenance Exclusion Windows AreConfiguring both instances on the Week 1 (canary) track causes both environments to receive updates 7 to 14 days after notification. Meanwhile, maintenance exclusion windows are a policy mechanism used in services like Google Kubernetes Engine (GKE) and Google Distributed Cloud connected to forbid automatic upgrades during specific non-repeating time periods, such as holiday sales freezes.
Week 1 causes it to update in the same 7 to 14 day window as staging rather than waiting 15 to 21 days.Week 1 track, production could be updated simultaneously with or even before staging, preventing proper compatibility validation.Using maintenance exclusion windows would be relevant if you were managing a GKE cluster and needed to temporarily block automatic control plane or node upgrades during a specific high-traffic retail event, rather than configuring recurring off-peak database maintenance in Cloud SQL.
In Cloud SQL, maintenance settings combine two core controls: a preferred maintenance window (day of the week and start hour in UTC) and a maintenance timing track (also known as the update order). Once you opt in to maintenance notifications, Cloud SQL sends an email alert when an update becomes available, and the selected track determines how many days later the instance receives the rollout.
Week 1 track (also labeled preview or canary) schedules its maintenance 7 to 14 days after the initial notification. This gives engineers advance notice before updates hit staging and enough time to execute functional and performance tests.Week 2 track (also labeled production or stable) schedules its update 15 to 21 days after notification. This creates a reliable buffer after staging updates while adopting new features well within the one-month limit.Week 1 and Week 2 provide guaranteed multi-day notification lead times.Compared to Any (which offers no predictable delay) or Week 5 (which delays updates for more than a month), pairing Week 1 for staging with Week 2 for production at Sunday 01:00 UTC delivers the exact balance of stability, testability, and timely feature adoption.
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