A reports an inadequate or failed acceptance condition: Worker instances remain at peak capacity and cost even when no customers call the service. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that capacity responds to workload while idle compute cost falls and latency remains acceptable. The underlying design requirement is serverless or autoscaled execution aligns spend with demand when stateless constraints are satisfied.

C reports an inadequate or failed acceptance condition: The Lambda workload hits timeout limits on real requests despite appearing cheap at idle. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that capacity responds to workload while idle compute cost falls and latency remains acceptable. The underlying design requirement is serverless or autoscaled execution aligns spend with demand when stateless constraints are satisfied.

D reports an inadequate or failed acceptance condition: Autoscaled containers never drop below their peak-sized minimum task allocation. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that capacity responds to workload while idle compute cost falls and latency remains acceptable. The underlying design requirement is serverless or autoscaled execution aligns spend with demand when stateless constraints are satisfied.

A would give every developer permanent full access to production secret values rather than runtime IAM roles. This is not a defensible answer to the stated constraint (A team wants to eliminate recurring outages caused by manually rotating database credentials): managed secrets lifecycle can replace error-prone manual distribution and rotation processes. The required design is integrate Secrets Manager rotation and application retrieval with runtime IAM roles; acceptance depends on evidence that rotation completes and application connections refresh without embedded long-lived passwords.

B would rotate the secret in Secrets Manager but never update application connection behavior to retrieve it. This is not a defensible answer to the stated constraint (A team wants to eliminate recurring outages caused by manually rotating database credentials): managed secrets lifecycle can replace error-prone manual distribution and rotation processes. The required design is integrate Secrets Manager rotation and application retrieval with runtime IAM roles; acceptance depends on evidence that rotation completes and application connections refresh without embedded long-lived passwords.

D would rotate the database password manually and bake the new literal value into every container image. This is not a defensible answer to the stated constraint (A team wants to eliminate recurring outages caused by manually rotating database credentials): managed secrets lifecycle can replace error-prone manual distribution and rotation processes. The required design is integrate Secrets Manager rotation and application retrieval with runtime IAM roles; acceptance depends on evidence that rotation completes and application connections refresh without embedded long-lived passwords.

A reports an inadequate or failed acceptance condition: All developers can retrieve the live database secret without approved runtime need. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that rotation completes and application connections refresh without embedded long-lived passwords. The underlying design requirement is managed secrets lifecycle can replace error-prone manual distribution and rotation processes.

B reports an inadequate or failed acceptance condition: Secrets Manager completes rotation, but long-running processes keep their now-invalid credentials. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that rotation completes and application connections refresh without embedded long-lived passwords. The underlying design requirement is managed secrets lifecycle can replace error-prone manual distribution and rotation processes.

C reports an inadequate or failed acceptance condition: A rotated password is stored in application configuration and exposed to the deployment team. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that rotation completes and application connections refresh without embedded long-lived passwords. The underlying design requirement is managed secrets lifecycle can replace error-prone manual distribution and rotation processes.

B would create an analytics destination with no change-capture or update pipeline to keep datasets fresh. This is not a defensible answer to the stated constraint (The data science group exports daily CSV snapshots from a production database to run trend analysis): purpose-built pipelines separate analytical demand from transactional workloads. The required design is introduce a change-stream or managed data pipeline into an analytical store appropriate to queries; acceptance depends on evidence that fresh analytical datasets appear without large read spikes on the production writer.

C would continue exporting full daily CSV dumps directly from the high-load production database. This is not a defensible answer to the stated constraint (The data science group exports daily CSV snapshots from a production database to run trend analysis): purpose-built pipelines separate analytical demand from transactional workloads. The required design is introduce a change-stream or managed data pipeline into an analytical store appropriate to queries; acceptance depends on evidence that fresh analytical datasets appear without large read spikes on the production writer.

D would run even larger analytical SQL scans on the OLTP writer during business hours. This is not a defensible answer to the stated constraint (The data science group exports daily CSV snapshots from a production database to run trend analysis): purpose-built pipelines separate analytical demand from transactional workloads. The required design is introduce a change-stream or managed data pipeline into an analytical store appropriate to queries; acceptance depends on evidence that fresh analytical datasets appear without large read spikes on the production writer.

A reports an inadequate or failed acceptance condition: An analytical destination exists but data remains a day old without change capture. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that fresh analytical datasets appear without large read spikes on the production writer. The underlying design requirement is purpose-built pipelines separate analytical demand from transactional workloads.

C reports an inadequate or failed acceptance condition: Daily full-database exports still spike CPU on the transactional writer. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that fresh analytical datasets appear without large read spikes on the production writer. The underlying design requirement is purpose-built pipelines separate analytical demand from transactional workloads.

D reports an inadequate or failed acceptance condition: Analytical queries against the primary OLTP database slow customer transactions. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that fresh analytical datasets appear without large read spikes on the production writer. The underlying design requirement is purpose-built pipelines separate analytical demand from transactional workloads.

A would rely on release notes and manual email notifications instead of tested versioned API contracts. This is not a defensible answer to the stated constraint (Developers manually track breaking API changes between replatformed services): aPI contracts protect independently deployed services from hidden coupling introduced during modernization. The required design is introduce versioned API contracts and automated compatibility testing as pipeline gates; acceptance depends on evidence that a breaking interface change fails the integration contract test before rollout.

B would use unversioned API schemas and let production consumers discover breaking fields on deploy. This is not a defensible answer to the stated constraint (Developers manually track breaking API changes between replatformed services): aPI contracts protect independently deployed services from hidden coupling introduced during modernization. The required design is introduce versioned API contracts and automated compatibility testing as pipeline gates; acceptance depends on evidence that a breaking interface change fails the integration contract test before rollout.

D would mark a pipeline green after unit tests alone, without cross-service compatibility contracts. This is not a defensible answer to the stated constraint (Developers manually track breaking API changes between replatformed services): aPI contracts protect independently deployed services from hidden coupling introduced during modernization. The required design is introduce versioned API contracts and automated compatibility testing as pipeline gates; acceptance depends on evidence that a breaking interface change fails the integration contract test before rollout.

A reports an inadequate or failed acceptance condition: A breaking API field rename passes release gates because no consumer contract test runs. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a breaking interface change fails the integration contract test before rollout. The underlying design requirement is aPI contracts protect independently deployed services from hidden coupling introduced during modernization.

B reports an inadequate or failed acceptance condition: Services publish incompatible schema versions without a deprecation or compatibility policy. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a breaking interface change fails the integration contract test before rollout. The underlying design requirement is aPI contracts protect independently deployed services from hidden coupling introduced during modernization.

C reports an inadequate or failed acceptance condition: The unit test suite passes while dependent service integration calls return errors after deployment. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a breaking interface change fails the integration contract test before rollout. The underlying design requirement is aPI contracts protect independently deployed services from hidden coupling introduced during modernization.

B would write unstructured log lines without request identifiers or alert routing. This is not a defensible answer to the stated constraint (A converted workload has no proactive measure of latency or errors across its new microservices): modernization without observability can multiply debugging complexity across service boundaries. The required design is instrument service traces, structured logs, and SLO-based alerts across the request path; acceptance depends on evidence that a synthetic fault is visible as a correlated trace and generates a service-level alert.

C would create dashboard panels that no one monitors and no synthetic fault can trigger an actionable alert. This is not a defensible answer to the stated constraint (A converted workload has no proactive measure of latency or errors across its new microservices): modernization without observability can multiply debugging complexity across service boundaries. The required design is instrument service traces, structured logs, and SLO-based alerts across the request path; acceptance depends on evidence that a synthetic fault is visible as a correlated trace and generates a service-level alert.

D would collect CPU averages per microservice but omit correlated request traces and customer latency SLOs. This is not a defensible answer to the stated constraint (A converted workload has no proactive measure of latency or errors across its new microservices): modernization without observability can multiply debugging complexity across service boundaries. The required design is instrument service traces, structured logs, and SLO-based alerts across the request path; acceptance depends on evidence that a synthetic fault is visible as a correlated trace and generates a service-level alert.

A reports an inadequate or failed acceptance condition: Unstructured logs lack correlation IDs, so a failing dependency cannot be followed end to end. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a synthetic fault is visible as a correlated trace and generates a service-level alert. The underlying design requirement is modernization without observability can multiply debugging complexity across service boundaries.

C reports an inadequate or failed acceptance condition: A synthetic outage happens without an alert despite dashboard charts being present. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a synthetic fault is visible as a correlated trace and generates a service-level alert. The underlying design requirement is modernization without observability can multiply debugging complexity across service boundaries.

D reports an inadequate or failed acceptance condition: Individual service CPU values look normal while customer request traces are missing. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that a synthetic fault is visible as a correlated trace and generates a service-level alert. The underlying design requirement is modernization without observability can multiply debugging complexity across service boundaries.

A would remove the legacy fallback paths before modern-route health and compatibility tests pass. This is not a defensible answer to the stated constraint (An enterprise must gradually replace a legacy web application without disrupting all clients in one release): incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages. The required design is use a strangler pattern with controlled routing of selected features to new services; acceptance depends on evidence that traffic tests show selected routes served by modern services while fallback paths still operate.

B would replace the entire legacy application in a single irreversible cutover rather than routing features incrementally. This is not a defensible answer to the stated constraint (An enterprise must gradually replace a legacy web application without disrupting all clients in one release): incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages. The required design is use a strangler pattern with controlled routing of selected features to new services; acceptance depends on evidence that traffic tests show selected routes served by modern services while fallback paths still operate.

D would deploy a new feature service but never route any client requests to it for validation. This is not a defensible answer to the stated constraint (An enterprise must gradually replace a legacy web application without disrupting all clients in one release): incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages. The required design is use a strangler pattern with controlled routing of selected features to new services; acceptance depends on evidence that traffic tests show selected routes served by modern services while fallback paths still operate.

A reports an inadequate or failed acceptance condition: Selected requests reach the new backend, but legacy fallback was already removed before its health validation. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that traffic tests show selected routes served by modern services while fallback paths still operate. The underlying design requirement is incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages.

B reports an inadequate or failed acceptance condition: New service replacement succeeds only after switching every client off the legacy system at once. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that traffic tests show selected routes served by modern services while fallback paths still operate. The underlying design requirement is incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages.

C reports an inadequate or failed acceptance condition: The new route is deployed but no application traffic is ever directed to its service. This result does not establish the specified outcome for the current architecture. The decisive test must demonstrate that traffic tests show selected routes served by modern services while fallback paths still operate. The underlying design requirement is incremental migration reduces big-bang cutover risk and allows old functionality to retire in stages.