Plat-Arch-202 Exam Results

The correct selection is D. A complete rollback must restore previously existing metadata and remove components introduced by the failed release. A pre-release baseline plus destructiveChanges.xml for newly added components gives a reproducible reverse deployment; simply redeploying the old metadata does not automatically delete new components. From a release perspective, approved functionality must be isolated from unfinished work and promoted through a predictable, auditable path. Release calendars, validation, stakeholder approval, rollback planning, and branch discipline reduce the risk that a technically correct change becomes an operational failure during production promotion. The rejected alternatives weaken release control by mixing development states, skipping required validation, or treating production promotion as an ad hoc administrative task. For this scenario, the decisive point is therefore the platform behavior represented by the selected option (D. Create a sandbox from production to take the backup of existing metadata. To roll back deployment, use destructivechanges.xml to delete new components and then Deploy again to production using metadata from this sandbox.). Applying that rule consistently keeps the implementation aligned with Salesforce-supported lifecycle practices and makes the resulting change easier to review, validate, and operate across environments.

Study Guide reference: Releasing — release management; branching and hotfixes; production readiness; rollback strategy; seasonal releases; approvals.

The correct selection is A, B. Because regulations prohibit production data in non-production, UC can either mask a production-derived Full sandbox or generate synthetic data with production-like distribution and volume. Both approaches support realistic performance behavior without exposing regulated values. From a testing perspective, the goal is deterministic evidence that the solution behaves correctly under the conditions that matter. Tests should isolate external dependencies, use controlled data, exercise positive and negative paths, and reproduce realistic volume or user behavior where required without making production the first place defects are discovered. The rejected choices either test the wrong layer, introduce live-system dependencies, or detect problems too late to provide the deterministic early feedback expected from a mature test strategy. For this scenario, the decisive point is therefore the platform behavior represented by the selected options (A. Perform data masking on full sandbox after a refresh.; B. Generate mock data that mimics production data shape and volume.). Applying that rule consistently keeps the implementation aligned with Salesforce-supported lifecycle practices and makes the resulting change easier to review, validate, and operate across environments.

Study Guide reference: Testing — Apex testing; test-data isolation; mocks; regression testing; performance/load testing; sandbox test strategy.

The correct selection is A. Checkmarx findings must be reviewed individually. Real vulnerabilities should be remediated; legitimate false positives should be documented with a technical justification and included in the AppExchange security-review submission rather than ignored or mechanically changed simply to produce a zero-warning report. From a build perspective, the implementation should be source-driven, reviewable, automatable, and modular. Engineering controls such as version control, package boundaries, static analysis, coding standards, and isolated development environments reduce defects early and make changes easier to understand, test, and promote. The alternatives do not create the same repeatable engineering control: they either postpone defect detection, couple teams unnecessarily, or bypass the package/source model that should govern the change. For this scenario, the decisive point is therefore the platform behavior represented by the selected option (A. Review the Checkmarx errors. If there is no need to fix, mark them as false positive and attach explanation, then submit.). Applying that rule consistently keeps the implementation aligned with Salesforce-supported lifecycle practices and makes the resulting change easier to review, validate, and operate across environments.

Study Guide reference: Building — Salesforce DX; package development; coding standards; static analysis; source control; secure development.

The correct selection is D. Static code analysis can automatically and consistently enforce naming, maintainability, security, and other coding standards across teams and repositories. It scales better than relying only on human review and can be integrated as a CI quality gate before code is merged. From a build perspective, the implementation should be source-driven, reviewable, automatable, and modular. Engineering controls such as version control, package boundaries, static analysis, coding standards, and isolated development environments reduce defects early and make changes easier to understand, test, and promote. The remaining options are weaker because they address symptoms after code is written instead of improving the build process through automation, modularity, and explicit development standards. For this scenario, the decisive point is therefore the platform behavior represented by the selected option (D. Use static code analysis to enforce coding standards.). Applying that rule consistently keeps the implementation aligned with Salesforce-supported lifecycle practices and makes the resulting change easier to review, validate, and operate across environments.

Study Guide reference: Building — Salesforce DX; package development; coding standards; static analysis; source control; secure development.