SF-PD2 Testing, Debugging, and Deployment Practice Test 3 — 15 Questions

Question 1

A developer writes a test class for an Apex service that enforces a custom validation rule: an Account cannot be saved without a ShippingPostalCode. The test method creates an Account, inserts it with a valid postal code, then attempts to update it to a blank postal code inside a try-catch block using Test.startTest() and Test.stopTest(). The test passes locally but fails in the production deployment validation step with 'System.DmlException: Insert failed'. Which approach should the developer use to make the test reliable across all environments?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Create test data that satisfies the validation rule, and use a separate test method to verify the rule behavior by attempting an invalid update and asserting the exception.. The reliable approach is to keep the service test focused on valid data and create a separate test for the validation rule's negative path. This decouples the test from environment-specific validation rules and ensures consistent behavior. It also follows Salesforce testing best practices by isolating concerns and avoiding conditional logic that depends on org configuration.

Answer

Insert the Account with a blank postal code first, then update it to a valid value, so the validation rule is not triggered.

Answer

Wrap the DML statement in a try-catch block and assert that a DmlException was thrown, without relying on validation rules.

Answer

Use Test.isRunningTest() inside the trigger or validation rule to bypass the rule during test execution.

Question 2

A development team uses a source-driven workflow with Salesforce CLI and a Git repository. A developer must deploy a subset of metadata to a sandbox that includes only a new custom object, its fields, and a permission set that grants access to those fields. Which command should the developer use to deploy only these components?

Accepted Answer

sf project deploy start --source-dir force-app/main/default/objects/MyObject__c --source-dir force-app/main/default/permissionsets/MyPermSet.permissionset-meta.xml --target-org MySandbox. Targeted source deployments in the sf CLI use project deploy start with one or more --source-dir paths that enumerate the exact components to publish. Combining the object directory with the specific permission set file deploys precisely the requested metadata to the sandbox, avoiding unrelated changes and keeping the deploy fast.

Answer

sf project deploy start --manifest manifest/package.xml --target-org MySandbox --ignore-conflicts

Answer

sf project convert source --root-dir force-app --output-dir mdapi --target-org MySandbox

Answer

sf project retrieve start --manifest package.xml --target-org MySandbox --deploy-dependent-components

Question 3

A developer is preparing to deploy a new Lightning Web Component (LWC) that includes an Apex controller. The deployment will be done using a change set. The developer needs to ensure that the Apex controller's test class meets all requirements for production deployment. Which two actions should the developer take? (Choose two.)

Accepted Answer

Ensure the test class covers at least 75% of the Apex controller's code.. To deploy an Apex controller to production, the test class must provide at least 75% coverage of the controller's code and include assertions that validate the controller's behavior. These two actions ensure that the code is adequately tested and meets Salesforce's deployment requirements. Other considerations like avoiding SeeAllData and efficient execution are good practices but not mandatory for this scenario.

Answer

Annotate the test class with @isTest and make it private.

Answer

Ensure the test class runs in under 10 minutes to avoid deployment timeouts.

Answer

Use @isTest(SeeAllData=true) to access real data and improve coverage.

Question 4

A developer is preparing to deploy a new Apex class and a Lightning web component to production. The deployment fails with the error: 'You can't change the API version of an installed package component.' The developer's package is not managed, but the org has a managed package installed that includes a class with the same name. Which action should the developer take to resolve the deployment error?

Accepted Answer

Rename the Apex class to avoid a naming collision with the managed package component.. The error occurs because the deployment attempts to modify a component that belongs to an installed managed package. Renaming the Apex class avoids the collision and allows the deployment to proceed. Managed package components are immutable, so the developer must choose a different name for the new class.

Answer

Change the API version of the new Apex class to match the managed package's API version.

Answer

Uninstall the managed package, deploy the new class, then reinstall the package.

Answer

Add the new class to a new managed package and deploy that package instead.

Question 5

A developer's Apex test class calls a method that enqueues a Queueable job which inserts Task records. The test asserts that the Tasks exist immediately after Test.stopTest() returns, but the assertion fails intermittently in the sandbox. The Queueable class contains no callouts. What is the most likely cause of the failing assertion?

Accepted Answer

The Queueable class enqueues another Queueable job, so the second job runs after stopTest completes.. Test.stopTest() is the boundary that synchronously runs asynchronous work queued before it, so any job added during that execution is left pending. When a Queueable job chains a second job, the chained work is not covered by the same stopTest call. Asserting on records created by the chained job fails because that job has not executed yet.

Answer

Queueable jobs cannot insert Task records in a test context because Task is a setup-adjacent object.

Answer

The test class did not call Test.startTest() before enqueuing the Queueable job.

Answer

The test class is missing @isTest(SeeAllData=true), which is required for asynchronous jobs to commit DML.

Question 6

A developer is using the Salesforce CLI to deploy metadata to a production org. The deployment includes a new Apex trigger and a modified Lightning Web Component. The developer wants to validate the deployment without committing changes and receive detailed test results. Which command should the developer use?

Accepted Answer

sfdx force:source:deploy --checkonly --testlevel RunLocalTests --targetusername production. The correct command uses --checkonly to validate without committing and --testlevel RunLocalTests to run local tests, which is required for production deployments. This provides detailed test results without affecting the org. Other options use invalid flags or inappropriate test levels.

Answer

sfdx force:source:deploy --validate --testlevel RunAllTestsInOrg --targetusername production

Answer

sfdx force:source:deploy --dry-run --testlevel NoTestRun --targetusername production

Answer

sfdx force:mdapi:deploy --checkonly --testlevel RunLocalTests --targetusername production

Question 7

A developer needs to write an Apex test class for a trigger that fires on Account insert. The trigger updates the Account's Rating field based on the AnnualRevenue. To ensure the test covers the trigger logic, which annotation must be used on the test method?

Accepted Answer

@testMethod. In Apex, test methods must be annotated with @testMethod. The class containing them must be annotated with @isTest. The @testMethod annotation tells the Salesforce platform that the method is a test method, allowing it to be executed during test runs and exempting it from governor limits. Without it, the method is not recognized as a test.

Answer

@TestSetup

Answer

@isTest

Answer

@AuraEnabled

Question 8

A developer is using Salesforce CLI to deploy metadata to a sandbox. The deployment command includes the --test-level RunLocalTests flag. After deployment, the developer notices that a test class that was previously passing now fails due to a CPU timeout. The test class performs a complex calculation and is not related to the deployed changes. Which action should the developer take to address the CPU timeout without compromising test coverage?

Accepted Answer

Move the CPU-intensive calculation into a Queueable job and test it asynchronously using Test.startTest() and Test.stopTest().. Moving CPU-intensive logic into a Queueable job allows it to run in a separate transaction with its own CPU limit. Testing it with Test.startTest() and Test.stopTest() ensures the job executes synchronously within the test, capturing results. This resolves the timeout without sacrificing coverage.

Answer

Break the test class into smaller test methods and use Test.startTest() and Test.stopTest() around each critical section to distribute CPU usage.

Answer

Increase the CPU timeout limit for the test class using the @Test annotation.

Answer

Refactor the test class to use Test.startTest() and Test.stopTest() to isolate the CPU-intensive logic and reduce the overall CPU time.

Question 9

A developer needs to debug an Apex class that is throwing a NullPointerException during a complex business process. The developer wants to capture detailed execution logs, including variable assignments and method entry/exit, for a specific user. Which tool should the developer use to obtain these logs?

Accepted Answer

Developer Console's Logs tab with debug levels set to FINEST for Apex Code and Apex Profiling.. The Developer Console's Logs tab with FINEST debug levels provides the most detailed execution logs, including variable assignments and method calls, which are essential for debugging a NullPointerException. Other options either do not provide code execution logs or would not capture any logs due to disabled logging.

Answer

Debug Logs in Setup with the user's trace flag set to NONE.

Answer

Apex Replay Debugger in Visual Studio Code with a checkpoint set on the user.

Answer

Setup Audit Trail to review recent configuration changes.

Question 10

A developer needs to run Apex tests for a deployment but wants to exclude tests that are known to be flaky and unrelated to the changed metadata. The team uses Salesforce CLI in a CI pipeline. Which two approaches allow the developer to control exactly which tests run during validation? (Choose two.)

Accepted Answer

Create a test suite in the org and pass its name to the sf apex run test command with the --suite-names flag.. Selective test execution in CI is achieved through the Salesforce CLI test commands. Naming individual classes with --tests or invoking a stored test suite with --suite-names both limit execution to a defined set, which addresses flaky and irrelevant tests while keeping the deployment validation meaningful. Both are supported, scriptable options for pipeline automation.

Answer

Run sf project deploy validate with the --ignore-errors flag to bypass failing test classes.

Answer

Set the ApexTestExecution limit in Setup to disable the offending test classes org-wide.

Answer

Add the @isTest(IsParallel=false) annotation to flaky test classes so the platform skips them during validation.

Question 11

A developer is preparing a second-generation managed package for release. The package must expose only a specific subset of Apex classes and Visualforce pages to subscriber orgs, and the developer wants to confirm the package contents before uploading a new version. Which metadata type should the developer inspect to verify that only intended components are included in the package?

Accepted Answer

The package manifest (package.xml) embedded in the second-generation package version. Second-generation managed packages capture their contents at version creation, and the embedded manifest is the definitive record of what is included. Reviewing it before promotion lets the developer confirm that only intended Apex classes and Visualforce pages are part of the version and catch accidental inclusions early, rather than discovering them after subscribers install the package.

Answer

The package's Apex class access and Visualforce page access settings in the package definition file

Answer

The sfdx-project.json packageDirectories and packageAliases entries

Answer

package.xml manifest used with the Metadata API

Question 12

A developer is deploying metadata to a production org using the Salesforce CLI. The deployment must create a custom object, deploy its fields, and then run Apex tests, but the developer wants each component type validated in a specific order and wants a pre-deployment destructive change to remove an obsolete field. Which capability directly supports this need?

Accepted Answer

A destructiveChanges.xml file paired with the main manifest, passed through the sf project deploy start command.. Removing an obsolete field while deploying new components is a destructive change, which the Metadata API handles through a dedicated destructiveChanges.xml manifest submitted together with the constructive manifest. The CLI exposes this through sf project deploy start, and the platform then processes creations, updates, and deletions internally while running the requested Apex tests.

Answer

A post-deployment script defined in the sfdx-project.json configuration that issues a deleteMetadata call.

Answer

A .forceignore file in the project root that excludes the obsolete field from the deployment.

Answer

A package.xml manifest that lists the components in the desired deployment sequence.

Question 13

A developer is preparing to deploy a new Apex class and a Lightning web component from a sandbox to production using a change set. The developer wants to ensure the deployment succeeds and that the new functionality works as expected after deployment. (Choose two.)

Accepted Answer

Run all local Apex tests in the sandbox and confirm that code coverage for the new Apex class meets the 75% requirement. Before deploying, running local Apex tests confirms that the new class meets the 75% coverage requirement and that existing tests still pass. The change set must then include the Apex class, its test class, and the Lightning web component bundle so that production validation can run the tests and the component is available. Together these steps ensure the deployment succeeds and the functionality is present after deployment.

Answer

Set the deployment test level to RunLocalTests in the change set deployment options to run all tests in the production org

Answer

Deploy the change set without running tests, then run the new Apex class's tests manually in production to verify coverage

Answer

Use the Force.com Migration Tool to retrieve the metadata from the sandbox and deploy it to production instead of using a change set

Question 14

A developer is investigating why an Apex test method intermittently fails in a continuous integration pipeline but passes locally. The test asserts on records created by a Queueable job that chains another Queueable. Which TWO practices should the developer apply to make the test deterministic? (Choose two.)

Accepted Answer

Ensure the test creates its own data with @TestSetup or within the method, then use Test.stopTest before asserting on the chained job's results.. Deterministic coverage of a Queueable chain requires forcing the asynchronous work to complete inside the test transaction and removing dependence on ambient org data. The Test.startTest and Test.stopTest boundary executes queued jobs synchronously so assertions run after the chain finishes, while self-contained data creation keeps results stable across orgs and pipeline runs.

Answer

Increase the Apex test timeout by adding @isTest(seeAllData=true) to the test class so the asynchronous job has time to finish.

Answer

Use Test.getEventBus and Test.enqueueJob to control scheduling of the queued jobs within the test.

Answer

Call Test.isRunningTest inside the Queueable to skip the chained second job during tests.

Question 15

A developer needs to inspect the debug logs for a specific user who is experiencing intermittent errors in a Lightning Web Component that calls an Apex controller. The developer wants to capture logs only for that user and only for the Apex class involved, without generating logs for other users or classes. Which tool should the developer use to configure a user-specific and class-specific trace flag?

Accepted Answer

Setup's Debug Logs page. Setup's Debug Logs page is the central place to manage trace flags. You can add a trace flag for a specific user, set a debug level that includes Apex code, and apply a class filter to limit logging to a particular Apex class. This precisely matches the requirement to capture logs for one user and one class. Other tools either lack the configuration capability or are for different purposes.

Answer

Apex Replay Debugger in VS Code

Answer

Salesforce CLI command 'sf apex log get'

Answer

Developer Console's Logs tab