AZ-204 Develop Azure compute solutions Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Use Durable Functions for orchestration.

  Why it's wrong here

  Durable Functions are designed for implementing stateful workflows and long-running orchestrations, which introduce significant complexity and overhead. For reliably processing individual events from Event Grid, their advanced capabilities like chaining functions or managing external interactions are unnecessary. Event Grid subscriptions inherently provide mechanisms for reliable delivery and error handling without requiring a separate orchestration layer, making Durable Functions an over-engineered solution here.

• ✓

  Enable retry policy on the Event Grid subscription.

  Why this is correct

  Enabling a retry policy on the Event Grid subscription is a fundamental mechanism for handling transient failures when delivering events to the Azure Function. This policy automatically reattempts delivery to the endpoint if the function returns an HTTP error (e.g., 4xx or 5xx), ensuring that temporary issues like network glitches or service unavailability do not result in lost events. Configuring appropriate retry attempts and backoff intervals significantly enhances the reliability of event processing.

• ✗

  Implement manual checkpointing in the function code.

  Why it's wrong here

  Implementing manual checkpointing in the function code is entirely inappropriate for processing events from Event Grid. Checkpointing is a concept specific to Event Hubs and Kafka, where consumers explicitly track their progress within a partition to ensure at-least-once delivery and enable recovery from failures. Event Grid manages event delivery and acknowledgment automatically, abstracting away the need for consumer-side progress tracking and making manual checkpointing irrelevant.

• ✓

  Configure a dead-letter destination for undelivered events.

  Why this is correct

  Configuring a dead-letter destination, typically an Azure Storage Blob container, is crucial for handling events that cannot be successfully delivered or processed even after exhausting all retry attempts. This mechanism ensures that events causing persistent errors are not lost but are instead moved to a designated location for manual inspection, debugging, or reprocessing. Dead-lettering provides a robust safety net, preventing data loss for problematic events and enabling comprehensive error management.

• ✗

  Use a queue trigger instead of an Event Grid trigger.

  Why it's wrong here

  Using a queue trigger instead of an Event Grid trigger would fundamentally alter the eventing architecture and introduce unnecessary complexity for this scenario. While queues are excellent for decoupling and reliable messaging, Event Grid is specifically designed for reactive, push-based event distribution from various Azure services. Introducing a queue between Event Grid and the function would require an additional component to forward events, negating Event Grid's direct integration benefits and increasing operational overhead.

Go deeper

Related to this question