View Full Salesforce Certified Integration Architect Exam Dumps and Practice Test Dumps.

Q201. Which pattern is best when Salesforce must receive status updates from a long running external process

1. Status callback
2. Full data replication
3. Browser polling
4. Report subscription

Correct Answer: 1. Status callback

Explanation

A status callback allows the external process to notify Salesforce when its state changes or when processing completes. Full data replication copies information rather than reporting progress. Browser polling creates repeated requests and depends on an active user session. Report subscriptions are analytics features. Status callback is therefore the correct answer because long running work should usually proceed independently while Salesforce receives updates only when meaningful changes occur. Architects should include correlation identifiers authentication retry handling and duplicate protection so status messages can be matched reliably to the original request.

Q202. Which control helps prevent stale data from remaining indefinitely in an integration cache

1. Shared credentials
2. Unlimited retention
3. Cache expiration policy
4. Disable monitoring

Correct Answer: 3. Cache expiration policy

Explanation

A cache expiration policy defines how long cached information remains valid before it must be refreshed or removed. Shared credentials do not manage freshness. Unlimited retention can preserve outdated values indefinitely. Disabling monitoring reduces visibility. Cache expiration policy is therefore the correct answer because cached reference or lookup data should balance performance with accuracy. Architects should select expiration periods according to how frequently source data changes and how much stale information the business can tolerate. Event based invalidation can also be used when changes need to be reflected more quickly.

Q203. Which approach best protects a Salesforce integration when an external payload contains unexpected extra fields

1. Fail every message automatically
2. Tolerant reader design
3. Remove schema validation
4. Share internal object structure

Correct Answer: 2. Tolerant reader design

Explanation

A tolerant reader design allows consumers to ignore additional fields they do not understand while still validating the fields they depend on. Failing every message for harmless additions makes interfaces fragile. Removing schema validation can allow invalid required data. Sharing internal object structure increases coupling. Tolerant reader design is therefore the correct answer because integration contracts often evolve by adding optional fields. Existing consumers should continue operating as long as the meaning of required fields remains unchanged. Breaking changes still require versioning and coordinated migration.

Q204. Which Salesforce feature supports receiving event messages through event based APIs

1. Dashboard folders
2. Static Resources
3. Approval Processes
4. Pub Sub API

Correct Answer: 4. Pub Sub API

Explanation

Pub Sub API provides an event based interface for publishing and subscribing to Salesforce event messages. Dashboard folders organize analytics. Static Resources store files. Approval Processes manage approvals. Pub Sub API is therefore the correct answer because external applications can consume platform events and change events without repeatedly polling record data. Architects should consider subscriber checkpoints replay behavior throughput authentication and error handling. Event consumers should also be designed for duplicate tolerance and recovery after temporary outages.

Q205. Which integration design best supports an external consumer that requires only records changed since the last successful synchronization

1. Incremental extraction
2. Full export every time
3. Manual file copy
4. Browser automation

Correct Answer: 1. Incremental extraction

Explanation

Incremental extraction retrieves only records that changed after a known successful synchronization point. Full exports repeatedly move unchanged information and consume more processing capacity. Manual copying is difficult to control. Browser automation is fragile for enterprise integration. Incremental extraction is therefore the correct answer because it reduces API usage data transfer and downstream processing. Architects should maintain reliable checkpoints and consider overlapping windows or reconciliation to avoid missing records near timing boundaries or after partial failures.

Q206. Which architecture pattern provides one controlled entry point for several integration services

1. Shared database
2. API facade
3. Direct endpoint exposure
4. Browser session reuse

Correct Answer: 2. API facade

Explanation

An API facade provides a stable controlled interface in front of one or more backend services. Shared databases create tight coupling. Direct exposure of every internal endpoint can increase security and maintenance complexity. Browser session reuse is inappropriate for service integration. API facade is therefore the correct answer because it can hide backend implementation details and centralize routing security or versioning. Architects should keep the facade focused and avoid turning it into an oversized layer containing unrelated business logic.

Q207. Which message property helps determine whether an integration message is too old to process safely

1. Record owner
2. User profile
3. Message timestamp
4. Dashboard name

Correct Answer: 3. Message timestamp

Explanation

A message timestamp indicates when an event or request was created and can help determine whether it is still valid for processing. Record ownership and user profiles do not indicate message age. Dashboard names are unrelated. Message timestamp is therefore the correct answer because delayed messages may no longer represent current business state. Architects should define expiration rules for time sensitive transactions and distinguish between messages that should still be processed and those that should be rejected or reconciled after excessive delay.

Q208. Which security design is best when different integrations require different Salesforce permissions

1. One administrator account for everything
2. Anonymous access
3. Shared credentials
4. Separate least privilege identities

Correct Answer: 4. Separate least privilege identities

Explanation

Separate least privilege identities give each integration only the permissions needed for its specific responsibilities. A shared administrator account grants excessive access and makes actions difficult to trace. Anonymous access lacks accountability. Shared credentials also reduce traceability and complicate revocation. Separate least privilege identities is therefore the correct answer because each integration can be monitored independently and restricted to the objects fields and operations it actually requires. Access should be reviewed periodically as responsibilities change.

Q209. Which integration technique is best when several external operations must either all complete or be logically reversed

1. Distributed saga
2. Dashboard refresh
3. File export
4. Static resource deployment

Correct Answer: 1. Distributed saga

Explanation

A distributed saga coordinates a sequence of local transactions and defines compensation when later steps fail. Dashboard refreshes and file exports do not manage distributed business transactions. Static resource deployment is unrelated. Distributed saga is therefore the correct answer because independent systems usually cannot participate in one shared database transaction. Architects should define each step its compensation state tracking and retry behavior. Compensation actions should also be idempotent because they may themselves need to be retried after partial failures.

Q210. Which monitoring value best indicates that an event consumer is falling behind

1. User count
2. Processing lag
3. Dashboard size
4. Report count

Correct Answer: 2. Processing lag

Explanation

Processing lag measures the delay between when an event becomes available and when a consumer successfully processes it. User count dashboard size and report count do not measure event backlog. Processing lag is therefore the correct answer because queue depth alone may not show how late important transactions have become. Architects should monitor lag together with throughput failure rate and queue depth. Increasing lag can indicate insufficient consumer capacity repeated errors or an unavailable downstream dependency.

Q211. Which design best handles an external service that returns different error codes for temporary and permanent failures

1. Retry every error forever
2. Ignore all errors
3. Error specific recovery rules
4. Delete failed requests

Correct Answer: 3. Error specific recovery rules

Explanation

Error specific recovery rules classify failures and apply the appropriate response to each category. Temporary service errors may be retried while validation errors usually require correction. Retrying everything indefinitely wastes resources. Ignoring errors creates data gaps. Deleting failed requests risks data loss. Error specific recovery rules is therefore the correct answer because resilient integrations must distinguish recoverable infrastructure problems from permanent business errors. Architects should document classifications retry limits escalation ownership and final failure handling.

Q212. Which architecture is best when Salesforce must send the same business event to several regions independently

1. One synchronous chain
2. Manual forwarding
3. Shared browser session
4. Regional event subscribers

Correct Answer: 4. Regional event subscribers

Explanation

Regional event subscribers allow each region to consume and process the same business event independently. A synchronous chain creates latency and availability dependencies. Manual forwarding does not scale. Shared browser sessions are not suitable for backend integration. Regional event subscribers is therefore the correct answer because each region can scale recover and apply local processing without blocking others. Architects should consider event retention ordering region specific filtering and how global identifiers are preserved across all subscribers.

Q213. Which practice helps ensure transformed currency values use consistent exchange rules across systems

1. Centralized conversion rules
2. Let each system guess the rate
3. Use text labels only
4. Ignore currency codes

Correct Answer: 1. Centralized conversion rules

Explanation

Centralized conversion rules define how currency values are interpreted converted and rounded across participating systems. Allowing each system to choose its own rate creates inconsistency. Text labels alone are insufficient. Ignoring currency codes can produce serious business errors. Centralized conversion rules is therefore the correct answer because integrations involving multiple currencies need a consistent source of rates effective dates and rounding behavior. Architects should also identify the authoritative system and preserve original currency information when required for audit or reconciliation.

Q214. Which API design principle helps consumers request only the fields they require

1. Maximum payload design
2. Field selection
3. Shared administrator access
4. Full object replication

Correct Answer: 2. Field selection

Explanation

Field selection allows consumers to request or receive only the data elements needed for their use case. Maximum payloads and full replication move unnecessary information. Shared administrator access creates excessive privileges. Field selection is therefore the correct answer because smaller payloads reduce network usage processing cost and unnecessary exposure of sensitive information. Architects should design interfaces so optional fields can be selected where appropriate while still keeping contracts understandable and avoiding excessive complexity.

Q215. Which Salesforce integration option supports custom business logic exposed through SOAP

1. Bulk API
2. Change Data Capture
3. Apex web service
4. Static Resource

Correct Answer: 3. Apex web service

Explanation

An Apex web service can expose custom Salesforce business logic through SOAP operations. Bulk API is designed for high volume record processing. Change Data Capture publishes record change events. Static Resources store files. Apex web service is therefore the correct answer when a SOAP consumer needs a custom operation not available through standard APIs. Architects should prefer standard interfaces when possible because custom services require additional development security testing documentation and governor limit management.

Q216. Which design is best when different consumers need separate retry policies for the same event stream

1. One common retry counter
2. Shared transaction state
3. Global synchronous processing
4. Consumer specific retry handling

Correct Answer: 4. Consumer specific retry handling

Explanation

Consumer specific retry handling allows each subscriber to respond to failures according to its own business and technical requirements. A shared retry counter or global synchronous process couples unrelated consumers. Shared transaction state can make independent recovery difficult. Consumer specific retry handling is therefore the correct answer because one subscriber may tolerate long delays while another requires quick retries or immediate escalation. Architects should still define common observability and governance so retry behavior remains understandable across the integration landscape.

Q217. Which practice best ensures messages can be traced back to their original source transaction

1. Preserve source transaction ID
2. Generate unrelated IDs at every step
3. Remove message metadata
4. Disable logging

Correct Answer: 1. Preserve source transaction ID

Explanation

Preserving the source transaction ID allows downstream systems and middleware to connect messages back to the originating business action. Generating unrelated identifiers at every stage makes tracing difficult. Removing metadata and disabling logging reduce operational visibility. Preserve source transaction ID is therefore the correct answer because end to end troubleshooting and reconciliation depend on stable identifiers. Additional correlation identifiers may be added but the original business reference should remain available where appropriate so technical records can be connected with business transactions.

Q218. Which mechanism is most appropriate when an external system must confirm message receipt quickly but process it later

1. Keep the connection open until completion
2. Immediate acknowledgment
3. Delete the message after receipt
4. Block the sender

Correct Answer: 2. Immediate acknowledgment

Explanation

An immediate acknowledgment confirms that the receiving system accepted responsibility for the message while actual business processing continues asynchronously. Keeping the connection open creates unnecessary latency and timeout risk. Deleting the message can cause data loss. Blocking the sender prevents valid integration traffic. Immediate acknowledgment is therefore the correct answer because it separates transport acceptance from final business completion. Architects should define what the acknowledgment means and provide a separate mechanism for communicating final success or failure when the sender needs that information.

Q219. Which integration concern should be reviewed when large payloads contain nested related records

1. Payload complexity
2. Dashboard ownership
3. Browser preference
4. Report folder count

Correct Answer: 1. Payload complexity

Explanation

Payload complexity affects validation transformation processing time and error handling when messages contain many nested related records. Dashboard ownership browser preferences and report folder counts are unrelated. Payload complexity is therefore the correct answer because deeply nested structures can increase coupling and make partial error recovery more difficult. Architects should consider whether the payload should be simplified divided into smaller messages or processed through an orchestration layer. The chosen structure should remain understandable and compatible with downstream system limits.

Q220. Which governance practice ensures obsolete integration endpoints are formally retired

1. Keep every endpoint forever
2. Share all credentials permanently
3. Endpoint decommissioning process
4. Disable documentation

Correct Answer: 3. Endpoint decommissioning process

Explanation

An endpoint decommissioning process defines how obsolete integration interfaces are identified communicated disabled and removed safely. Keeping every endpoint indefinitely increases security and maintenance burden. Permanent shared credentials create unnecessary risk. Disabling documentation makes dependencies harder to understand. Endpoint decommissioning process is therefore the correct answer because old integrations may still have unknown consumers or credentials. Architects should inventory usage notify owners define retirement dates monitor remaining traffic and revoke associated access once the endpoint is no longer required.