Google Associate Cloud Engineer Practice Test 64
Question 630
A team defines a 99.9% availability SLO but never specifies how it counts failed user requests. What first definition is required?
- Use appropriate short- and long-window error-budget burn rate conditions for the documented service-level objective.
- Disable the alerting policy so the current errors no longer appear.
- Rename the monitored VM and assume metric-series selection is automatically corrected.
- Define the service-level indicator's eligible requests, good responses, measurement window and exclusion rules.
Explanation
Correct Answer
D:An SLO target is not measurable without a precise SLI definition and denominator.
Incorrect Answers
A:It does not measure or notify on the signal requested by the case. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Define the service-level indicator's eligible requests, good responses, measurement window and exclusion rules.
B:Removing an alert conceals the failing signal instead of resolving its cause. The specific requirement concerns the relevant monitoring signal or alert. The appropriate action here is: Define the service-level indicator's eligible requests, good responses, measurement window and exclusion rules.
C:Renaming a resource does not alter the actual effective configuration. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Define the service-level indicator's eligible requests, good responses, measurement window and exclusion rules.
Question 631
A service has exhausted its monthly error budget, but a team still plans an optional high-risk feature release with no mitigation. What should release governance consider?
- Use appropriate short- and long-window error-budget burn rate conditions for the documented service-level objective.
- Apply an approved error-budget policy that delays risky changes or requires explicit exception and mitigation.
- Set an unrelated CPU threshold without selecting the actual signal or metric labels.
- Disable the alerting policy so the current errors no longer appear.
Explanation
Correct Answer
B:SLO error-budget consumption provides a risk signal for releases; an incident rate can justify slowing discretionary change.
Incorrect Answers
A:It does not measure or notify on the signal requested by the case. The specific requirement concerns the relevant monitoring signal or alert. The appropriate action here is: Apply an approved error-budget policy that delays risky changes or requires explicit exception and mitigation.
C:It does not measure or notify on the signal requested by the case. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Apply an approved error-budget policy that delays risky changes or requires explicit exception and mitigation.
D:Removing an alert conceals the failing signal instead of resolving its cause. The case instead depends on the relevant monitoring signal or alert. The appropriate action here is: Apply an approved error-budget policy that delays risky changes or requires explicit exception and mitigation.
Question 632
An on-call rotation changes daily but alert notifications are hard-coded to one engineer's personal address. What operational integration is needed?
- Update and validate the alert policy's notification channels and incident escalation routing using an approved test.
- Rename the monitored VM and assume metric-series selection is automatically corrected.
- Send incidents to a maintained team escalation channel or pager rotation rather than a single static contact.
- Inspect a single average value and ignore distribution, errors and affected cohorts.
Explanation
Correct Answer
C:Escalation based on one person is fragile when availability and ownership change; incident routes should track rotations.
Incorrect Answers
A:It does not measure or notify on the signal requested by the case. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Send incidents to a maintained team escalation channel or pager rotation rather than a single static contact.
B:Renaming a resource does not alter the actual effective configuration. The case instead depends on the relevant monitoring signal or alert. The appropriate action here is: Send incidents to a maintained team escalation channel or pager rotation rather than a single static contact.
D:It does not measure or notify on the signal requested by the case. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Send incidents to a maintained team escalation channel or pager rotation rather than a single static contact.
Question 633
A team reviews only aggregated availability across all regions, missing a complete outage for one small regional customer cohort. What measurement should be added?
- Segment availability indicators by relevant region and critical customer cohort while retaining an overall objective.
- Attach safe release or revision attributes to telemetry and mark deployments on performance dashboards.
- Increase Cloud Logging retention as a replacement for the required alert or metric definition.
- Rename the monitored VM and assume metric-series selection is automatically corrected.
Explanation
Correct Answer
A:Aggregate performance can mask severe impact on a subgroup if most traffic originates elsewhere.
Incorrect Answers
B:Deploying a separate workload does not automatically configure the existing feature. The case instead depends on the relevant monitoring signal or alert.
C:Cloud Logging records telemetry, rather than directly enforcing the requested runtime behavior. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Segment availability indicators by relevant region and critical customer cohort while retaining an overall objective.
D:Renaming a resource does not alter the actual effective configuration. The specific requirement concerns the relevant monitoring signal or alert. The appropriate action here is: Segment availability indicators by relevant region and critical customer cohort while retaining an overall objective.
Question 634
A severity-one incident is repeatedly reopened after rollback because the team never records when alerts clear. What closure evidence should be required?
- Separate resource-level failure from a zone/regional or provider-wide service incident using authoritative scoped status and telemetry.
- Increase Cloud Logging retention as a replacement for the required alert or metric definition.
- Disable the alerting policy so the current errors no longer appear.
- Verify service-level indicators and correlated error/latency metrics recover within an agreed observation window before closing.
Explanation
Correct Answer
D:A deployment rollback action alone does not establish the service is healthy if dependent systems remain degraded.
Incorrect Answers
A:Placement changes do not automatically implement the required resource-level feature. It leaves the requested latency measurement unaddressed.
B:Cloud Logging records telemetry, rather than directly enforcing the requested runtime behavior. The specific requirement concerns the requested latency measurement.
C:Removing an alert conceals the failing signal instead of resolving its cause. The necessary distinction is the requested latency measurement.
Question 635
Gemini Cloud Assist identifies an idle-looking disk as a cost candidate, but the disk contains the only encrypted backup for a regulated service. What must be validated?
- Check resource ownership, backup policy, actual attachment and retention classification before any AI-assisted deletion.
- Compare the recommendation with current authoritative VM inventory, telemetry, ownership and recovery-role evidence, recording the observation times.
- Send every alert to all organization members instead of configuring intended notification channels.
- Inspect a single average value and ignore distribution, errors and affected cohorts.
Explanation
Correct Answer
A:An optimization recommendation without data-retention context can be unsafe even if utilization is low.
Incorrect Answers
B:It does not measure or notify on the signal requested by the case. The specific requirement concerns retention or recoverability controls.
C:It does not measure or notify on the signal requested by the case. The necessary distinction is retention or recoverability controls.
D:It does not measure or notify on the signal requested by the case. The case instead depends on retention or recoverability controls.
Question 636
An AI assistant suggests increasing a Cloud SQL replica count to address a write-lock problem despite evidence pointing to one blocking transaction. How should the recommendation be assessed?
- Inspect database span attributes, query timing and related slow-query evidence before changing the front end.
- Rename the monitored VM and assume metric-series selection is automatically corrected.
- Compare the proposed capacity remedy with query and lock telemetry and reject it unless it addresses the verified bottleneck.
- Send every alert to all organization members instead of configuring intended notification channels.
Explanation
Correct Answer
C:Generative suggestions must be checked against actual operational symptoms; scaling replicas does not necessarily fix primary transaction locks.
Incorrect Answers
A:It does not measure or notify on the signal requested by the case. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Compare the proposed capacity remedy with query and lock telemetry and reject it unless it addresses the verified bottleneck.
B:Renaming a resource does not alter the actual effective configuration. The case instead depends on the relevant monitoring signal or alert. The appropriate action here is: Compare the proposed capacity remedy with query and lock telemetry and reject it unless it addresses the verified bottleneck.
D:It does not measure or notify on the signal requested by the case. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Compare the proposed capacity remedy with query and lock telemetry and reject it unless it addresses the verified bottleneck.
Question 637
Gemini Cloud Assist recommends deleting several idle VMs. The answer references a resource inventory collected weeks ago, while the operations team needs a current risk decision. What should they verify before acting?
- Treat the AI response timestamp as proof of current VM state; no resource query is necessary if the text was generated today.
- Delete the VMs and use Cloud Billing credits if a recovery needs to be funded later.
- Compare the recommendation with current authoritative VM inventory, telemetry, ownership and recovery-role evidence, recording the observation times.
- Switch the assistant's display language; the model's language choice forces stale infrastructure data to refresh.
Explanation
Correct Answer
C:Compare the recommendation with current authoritative VM inventory, telemetry, ownership and recovery-role evidence, recording the observation times. Live resource state, relevant utilization metrics and business recovery ownership provide the evidence needed to test a stale suggestion safely.
Incorrect Answers
A:Treat the AI response timestamp as proof of current VM state; no resource query is necessary if the text was generated today. Generated response time differs from when the cited infrastructure snapshot was captured. It cannot establish the VMs are still idle or dispensable.
B:Delete the VMs and use Cloud Billing credits if a recovery needs to be funded later. Deleting without checking production dependencies can remove critical standby or recovery capacity; billing funds do not restore lost configuration.
D:Switch the assistant's display language; the model's language choice forces stale infrastructure data to refresh. Presentation language does not trigger a guaranteed refresh of inventory or replace a direct check against current cloud telemetry.
Question 638
An incident bot produces a suggested gcloud command that targets a broad organization scope while the incident concerns one sandbox project. What should the operator do?
- Check resource ownership, backup policy, actual attachment and retention classification before any AI-assisted deletion.
- Send every alert to all organization members instead of configuring intended notification channels.
- Narrow and review the command's resource scope and identity before executing it under normal change approval.
- Disable the alerting policy so the current errors no longer appear.
Explanation
Correct Answer
C:AI-assisted commands should be treated as editable proposals, never automatic authority to change unrelated environments.
Incorrect Answers
A:It does not measure or notify on the signal requested by the case. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Narrow and review the command's resource scope and identity before executing it under normal change approval.
B:It does not measure or notify on the signal requested by the case. The specific requirement concerns the relevant monitoring signal or alert. The appropriate action here is: Narrow and review the command's resource scope and identity before executing it under normal change approval.
D:Removing an alert conceals the failing signal instead of resolving its cause. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Narrow and review the command's resource scope and identity before executing it under normal change approval.
Question 639
A troubleshooting assistant omits a necessary rollback plan for a risky IAM policy edit. What should a production reviewer add?
- Inspect the service-specific audit logging defaults and effective project or organization policy before changing audit configuration.
- Disable the alerting policy so the current errors no longer appear.
- Increase Cloud Logging retention as a replacement for the required alert or metric definition.
- Document the exact prechange IAM policy, verification steps and supported rollback or recovery path before applying.
Explanation
Correct Answer
D:AI-generated remediation text cannot substitute for controlled reversibility and explicit access testing.
Incorrect Answers
A:Changing audit collection cannot grant a denied permission or fix service behavior. The specific requirement concerns the relevant monitoring signal or alert.
B:Removing an alert conceals the failing signal instead of resolving its cause. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Document the exact prechange IAM policy, verification steps and supported rollback or recovery path before applying.
C:Cloud Logging records telemetry, rather than directly enforcing the requested runtime behavior. The case instead depends on the relevant monitoring signal or alert. The appropriate action here is: Document the exact prechange IAM policy, verification steps and supported rollback or recovery path before applying.
Question 640
An AI-generated incident summary claims an outage started before the first recorded error event, but no independent evidence supports that timeline. What should be changed?
- Reconstruct start/end times from provider advisories, timestamped logs and service metrics rather than repeating unsupported summary text.
- Archive authorized provider event IDs, affected scopes, timestamps and local incident evidence in the incident record.
- Set an unrelated CPU threshold without selecting the actual signal or metric labels.
- Send every alert to all organization members instead of configuring intended notification channels.
Explanation
Correct Answer
A:Incident reports need evidence-based temporal claims; language-model summaries may infer timestamps that telemetry does not establish.
Incorrect Answers
B:It does not measure or notify on the signal requested by the case. The necessary distinction is the relevant monitoring signal or alert. The appropriate action here is: Reconstruct start/end times from provider advisories, timestamped logs and service metrics rather than repeating unsupported summary text.
C:It does not measure or notify on the signal requested by the case. The case instead depends on the relevant monitoring signal or alert. The appropriate action here is: Reconstruct start/end times from provider advisories, timestamped logs and service metrics rather than repeating unsupported summary text.
D:It does not measure or notify on the signal requested by the case. It leaves the relevant monitoring signal or alert unaddressed. The appropriate action here is: Reconstruct start/end times from provider advisories, timestamped logs and service metrics rather than repeating unsupported summary text.