Amazon AWS Certified Solutions Architect - Professional SAP-C02 Bundle

Administrators run sensitive maintenance scripts with no enforced approval or trail. The specific problem here is that engineers measure host utilization but not the outcomes experienced by users.

Capacity planning ignores demand spikes or hot keys that concentrate requests. The specific problem here is that engineers measure host utilization but not the outcomes experienced by users.

Config findings are left as unactioned notifications while risky drift remains exposed. The specific problem here is that engineers measure host utilization but not the outcomes experienced by users.

Approved application accounts can create authorized attachments to the centrally owned gateway. However, this does not establish the required outcome: dashboards identify targeted buckets with abnormal storage patterns and lifecycle candidates.

Tested resized instances meet p95 latency goals while recommended savings are observed. However, this does not establish the required outcome: dashboards identify targeted buckets with abnormal storage patterns and lifecycle candidates.

Finance can reconcile detailed usage line items to accounts and cost categories. However, this does not establish the required outcome: dashboards identify targeted buckets with abnormal storage patterns and lifecycle candidates.

Overlapping network prefixes make ordinary VPC peering route selection ambiguous. The stated evidence instead indicates: Events remain confined to local account buses because cross-account bus authorization is absent.

All accounts share one indiscriminate service-control policy despite different risk profiles. The stated evidence instead indicates: Events remain confined to local account buses because cross-account bus authorization is absent.

Each application team is provisioning its own disconnected backbone gateway. The stated evidence instead indicates: Events remain confined to local account buses because cross-account bus authorization is absent.

File-system throughput reaches target under representative concurrent processing without exhausted credits. However, this does not establish the required outcome: the repeated key lookup hits DAX with lower latency while noncacheable paths use the underlying table.

Clients use accelerator anycast addresses and connect to the healthy regional endpoint with improved path latency. However, this does not establish the required outcome: the repeated key lookup hits DAX with lower latency while noncacheable paths use the underlying table.

A poison message goes to the DLQ after configured attempts while valid following messages proceed. However, this does not establish the required outcome: the repeated key lookup hits DAX with lower latency while noncacheable paths use the underlying table.

Capacity planning ignores demand spikes or hot keys that concentrate requests. The specific problem here is that unsupported drivers or unsuitable instance networking features cap throughput.

Passive regional copies exist without health-driven endpoint selection. The specific problem here is that unsupported drivers or unsuitable instance networking features cap throughput.

Gp2 capacity-linked performance assumptions lead to wasteful storage growth. The specific problem here is that unsupported drivers or unsuitable instance networking features cap throughput.

Standard short-lived backups cannot recover the exact chosen accidental-update point. The stated evidence instead indicates: Teams confuse key material rotation with replacing the KMS key ID and losing access to old data.

Resource-based trust was not evaluated beyond local IAM identity permission reports. The stated evidence instead indicates: Teams confuse key material rotation with replacing the KMS key ID and losing access to old data.

WAF rules are deployed without visibility into which requests they affect. The stated evidence instead indicates: Teams confuse key material rotation with replacing the KMS key ID and losing access to old data.

Failed notifications disappear because no supported retry and DLQ design is in place. The specific problem here is that the application is exposed to one database host or zonal compute failure.

Only instance host metrics exist and individual container bottlenecks cannot be isolated. The specific problem here is that the application is exposed to one database host or zonal compute failure.

Health checks use an unrelated always-200 endpoint that does not reflect the service readiness. The specific problem here is that the application is exposed to one database host or zonal compute failure.

Dedicated EC2 servers remain the scaling bottleneck for stateless containerized workloads. The stated evidence instead indicates: All analytics jobs depend on a single custom ETL server and its directly attached storage.

Tight synchronous coupling propagates a noncritical downstream failure to the primary transaction. The stated evidence instead indicates: All analytics jobs depend on a single custom ETL server and its directly attached storage.

Migration estimates are based solely on outdated procurement records rather than measured workload behavior. The stated evidence instead indicates: All analytics jobs depend on a single custom ETL server and its directly attached storage.

Snapshot creation is mistaken for proof the application can actually be restored. The stated evidence instead indicates: The application is exposed to one database host or zonal compute failure.

Health checks use an unrelated always-200 endpoint that does not reflect the service readiness. The stated evidence instead indicates: The application is exposed to one database host or zonal compute failure.

Engineers measure host utilization but not the outcomes experienced by users. The stated evidence instead indicates: The application is exposed to one database host or zonal compute failure.

This action targets a different technical need: A build system hosted outside AWS needs temporary access to deploy CloudFormation without storing long-lived access keys. It does not replace either required solution in A or B. OIDC role federation supports temporary credentials and scoped trust for external workloads; verify the two stated requirements independently rather than accepting an unrelated change.

This action targets a different technical need: A public application must divert DNS traffic away from an endpoint when health checks show it has failed. It does not replace either required solution in A or B. DNS failover requires health-aware routing and realistic expectations for cached answers; verify the two stated requirements independently rather than accepting an unrelated change.

This action targets a different technical need: A global TCP service wants static public entry addresses and traffic steering over the AWS global network for lowest path latency. It does not replace either required solution in A or B. Global Accelerator accelerates supported TCP/UDP applications without serving as an HTTP content cache; verify the two stated requirements independently rather than accepting an unrelated change.

Eligible file changes are transferred and verified by a scheduled DataSync task. However, this does not establish the required outcome: partners connect over SFTP and transferred objects appear in assigned bucket prefixes.

Replicated servers boot in AWS test VPCs with verified application functionality before cutover. However, this does not establish the required outcome: partners connect over SFTP and transferred objects appear in assigned bucket prefixes.

Vendor-certified testing proves that the appliance runs on the selected target or requires retain/replace decision. However, this does not establish the required outcome: partners connect over SFTP and transferred objects appear in assigned bucket prefixes.