MULESOFT-PLATFORM-ARCHITECT-I Salesforce Certified MuleSoft Platform Architect I free practice test

All questions and explanations

Prefer to read at your own pace? Work through each question below, then open its answer to check your reasoning. You can link directly to a question or print this reference.

1. Question 1

   Link to question

   A platform architect is designing a new integration solution for a retail company. The requirement is to expose a customer-facing API that aggregates data from multiple backend systems and also provides a consistent interface for mobile and web applications. Which API-led connectivity layer should the architect use for the customer-facing API?

   1. System API

   2. Process API

   3. Experience API

   4. Connector API

   Show answer and explanation

   Correct answer

   • Experience API

   The Experience API layer is designed to present data in a format optimized for specific client channels, such as mobile or web applications. It aggregates data from Process and System APIs to provide a tailored interface. System APIs expose backend data, Process APIs orchestrate business processes, and Connector API is not a standard layer in API-led connectivity.

2. Question 2

   Link to question

   A MuleSoft team is deploying a critical application to CloudHub and wants to minimize downtime during releases. They have a single worker in production. Which deployment strategy best achieves minimal downtime without requiring additional infrastructure changes?

   1. Blue-green deployment with two separate environments

   2. Rolling update gradually replacing worker instances

   3. Canary deployment with traffic splitting on the same worker

   4. In-place redeployment with a brief maintenance window

   Show answer and explanation

   Correct answer

   • Blue-green deployment with two separate environments

   Blue-green deployment reduces downtime by maintaining two separate environments (blue and green) and switching traffic after the new version is fully ready. Rolling updates require multiple workers; canary deployment requires traffic splitting, which is not possible with a single worker; in-place redeployment causes downtime. Blue-green is the most effective with a single worker and minimal infrastructure.

3. Question 3

   Link to question

   During a complex integration flow, a Mule application uses an HTTP request to an external service that occasionally returns a 5xx error. The architect needs to ensure that when such an error occurs, the flow continues with a default value instead of failing. Which error handling strategy should the architect implement?

   1. Use on-error-continue in a flow reference component

   2. Use on-error-propagate in the error handler

   3. Wrap the HTTP request in a Try scope with on-error-continue

   4. Configure a global error handler with on-error-propagate

   Show answer and explanation

   Correct answer

   • Wrap the HTTP request in a Try scope with on-error-continue

   The Try scope allows handling errors locally without affecting the main flow. Setting on-error-continue inside the Try scope catches the error and lets the flow proceed with a default value. on-error-continue at flow level would skip the entire flow, on-error-propagate re-throws the error, and a global error handler would apply to all unhandled errors but still require local handling to provide a default.

4. Question 4

   Link to question

   A financial institution needs to secure a MuleSoft API that processes sensitive transaction data. The API must verify client identity and ensure that only authorized applications can invoke it. The security solution should also support rate limiting and token-based access. Which approach best meets these requirements?

   1. IP whitelisting with client credentials stored in secure properties

   2. HTTP Basic Authentication over TLS

   3. OAuth 2.0 with client credentials grant and rate limiting at the API gateway

   4. Client ID enforcement using Anypoint Platform's built-in policy

   Show answer and explanation

   Correct answer

   • OAuth 2.0 with client credentials grant and rate limiting at the API gateway

   OAuth 2.0 with client credentials grant provides token-based access and supports rate limiting centrally at the API gateway. It meets security and identity verification needs. IP whitelisting is too restrictive and doesn't handle tokens; Basic Auth is less secure; Client ID enforcement is a policy but lacks full OAuth capabilities like token expiration and refresh.

5. Question 5

   Link to question

   An operations team needs to monitor a Mule application running in CloudHub to proactively detect performance bottlenecks and errors. They want to set up alerts when the application's error rate exceeds a threshold over a 5-minute window. Which monitoring method should they use?

   1. Enable detailed logging in the application and review logs using Anypoint Monitoring logs

   2. Create a CloudHub alert based on the 'Error Count' metric with a 5-minute aggregation

   3. Use Anypoint Monitoring custom dashboards with real-time error rate visualization

   4. Set up an external application performance monitoring tool and integrate via HTTP request

   Show answer and explanation

   Correct answer

   • Create a CloudHub alert based on the 'Error Count' metric with a 5-minute aggregation

   CloudHub alerts can be configured directly on metrics like 'Error Count' with a specified aggregation window (e.g., 5 minutes) to trigger notifications. This is the simplest and native method for proactive alerting. Logs and dashboards are for analysis, not alerts; external tools add complexity and latency.

6. Question 6

   Link to question

   A company requires integrating a legacy inventory system that updates its data only when a specific event occurs. The system does not expose an API but can write to a file share. The inventory record count is moderate. Which integration approach is most appropriate?

   1. Poll the file share every minute using a Mule polling scheduler

   2. Use Anypoint MQ to asynchronously receive messages when a new file is written

   3. Set up a webhook listener on the file share to trigger a Mule flow

   4. Use a polling approach with a watermark to detect new files only

   Show answer and explanation

   Correct answer

   • Use Anypoint MQ to asynchronously receive messages when a new file is written

   Anypoint MQ provides event-driven integration where the legacy system can publish messages when new files are available, avoiding constant polling. Polling is resource-intensive and not event-driven; webhook is not natively supported on file shares; polling with watermark still polls but reduces data processing.

7. Question 7

   Link to question

   A Mule application processes high-volume transactions and is experiencing performance degradation due to thread contention. The application uses a non-blocking HTTP connector to receive requests. Which change would most effectively improve throughput for this scenario?

   1. Increase the 'Max Active' threads in the HTTP listener configuration

   2. Reduce the 'Connection Idle Timeout' to recycle threads faster

   3. Set the 'Processing Strategy' to 'custom' and increase the 'Max Concurrency'

   4. Use a 'Blocking' processing strategy and increase the thread pool size

   Show answer and explanation

   Correct answer

   • Set the 'Processing Strategy' to 'custom' and increase the 'Max Concurrency'

   For non-blocking HTTP, increasing 'Max Concurrency' allows more concurrent requests to be processed without additional threads, reducing thread contention. Increasing 'Max Active' threads may help but can cause more context switching; reducing timeout has minimal impact; blocking strategy would exacerbate contention.

8. Question 8

   Link to question

   An architect is designing a Mule application that must be deployed in a cluster across two CloudHub workers for high availability. The application uses an HTTP listener to receive requests. How should the architect configure the HTTP listener to properly load balance traffic across both workers?

   1. Configure the HTTP listener with a shared load balancer URL provided by CloudHub

   2. Use a private IP address in the listener configuration to restrict traffic to internal VPC

   3. Set the HTTP listener to run only on the primary worker and use a secondary as standby

   4. Define a custom DNS round-robin pointing to both worker IPs

   Show answer and explanation

   Correct answer

   • Configure the HTTP listener with a shared load balancer URL provided by CloudHub

   CloudHub automatically provides a load-balanced endpoint for clustered applications, distributing requests across workers. Using private IPs or custom DNS round-robin is unnecessary and not typically supported. Defining a primary/standby pattern defeats the purpose of clustering for high availability.

9. Question 9

   Link to question

   A multinational corporation requires its MuleSoft integrations to remain available even if an entire AWS region fails. The solution must handle failover automatically with minimal data loss. Which deployment model should the architect recommend?

   1. Active-Active deployment across two AWS regions using CloudHub's multi-site feature

   2. Active-Passive deployment with a single CloudHub worker in a secondary region

   3. On-premises Mule Runtime cluster with disaster recovery site

   4. CloudHub with auto-scaling at the primary region only

   Show answer and explanation

   Correct answer

   • Active-Active deployment across two AWS regions using CloudHub's multi-site feature

   Active-Active across two AWS regions provides automatic failover and minimal data loss as traffic can be split between regions. Active-Passive may have longer failover time; on-premises is not as resilient; auto-scaling only in the primary region does not protect against regional failure.

10. Question 10

    Link to question

    A project requires integrating a CRM system with an ERP system. The CRM sends a list of updated customer records every hour, and the ERP must receive the latest versions. Data volumes are moderate and near-real-time updates are acceptable. Which integration approach is most suitable?

    1. Use batch processing with MuleSoft's Batch module to process the entire list each hour

    2. Implement a real-time streaming flow that processes each record as it arrives

    3. Use a scheduled flow to poll the CRM every hour and upsert records into the ERP

    4. Leverage Anypoint MQ with a one-way message for each update

    Show answer and explanation

    Correct answer

    • Use a scheduled flow to poll the CRM every hour and upsert records into the ERP

    A scheduled flow polling the CRM every hour aligns with the CRM's batch update frequency and is simpler than real-time streaming or MQ. The Batch module adds overhead for simple upserts, and real-time streaming would require an event-driven trigger not provided by the CRM.