AWS Certified Solutions Architect Professional SAP-C02 Practice Question
Your fintech company runs an event-driven architecture on AWS. The Order service publishes every incoming order to an Amazon SNS FIFO topic named Orders.fifo. Several microservices-each subscribed through either an Amazon SQS FIFO queue or an AWS Lambda function-process these order events in parallel. During a recent incident one subscriber was down for 6 hours and could not easily retrieve the missed events after it came back online. The reliability team has issued the following additional requirements:
Each subscriber must be able to re-deliver (replay) up to the last 24 hours of order events on demand without help from the publisher.
Strict ordering and exactly-once message delivery must be preserved.
The solution must be fully managed and require only minimal configuration changes (no code changes) to existing publishers or consumers.
Which action will most effectively meet these requirements?
Subscribe an Amazon Kinesis Data Streams stream to the SNS topic and enable extended data retention; have subscribers read from the stream starting at a chosen sequence number when a replay is required.
Add an Amazon SQS Standard queue with a 24-hour retention period as an additional subscriber and configure consumers to redrive messages from the queue's dead-letter queue after an outage.
Enable message archiving on the existing SNS FIFO topic with a 24-hour retention period, and allow subscribers to start a replay by setting the ReplayPolicy attribute when they need to recover.
Turn on AWS CloudTrail data events for the SNS topic and use AWS Glue to extract the Publish API calls from the CloudTrail logs in Amazon S3; republish the messages when a subscriber requests a replay.
Amazon SNS FIFO topics include a managed message-archiving and replay feature. By enabling an archive policy on the topic and setting a 24-hour retention period, SNS automatically stores every published message in a highly durable, multi-AZ repository. Any subscriber can later set the ReplayPolicy attribute on its subscription to replay some or all archived messages. No code changes are required-only topic- and subscription-level configuration-and FIFO semantics (strict ordering and exactly-once delivery) remain intact.
The other options fail to meet one or more requirements:
CloudTrail records only API metadata, not the message payload, so the application data needed for replay would be missing.
Amazon SNS can fan out directly to Kinesis Data Firehose, but not to Kinesis Data Streams; even if it could, consumers would have to be rewritten to use the Kinesis API.
Adding an SQS Standard queue would break strict ordering and does not provide an after-the-fact replay for messages that were already acknowledged or deleted.
Therefore, enabling SNS FIFO message archiving and subscriber-initiated replay is the most effective and least intrusive solution.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
What is the managed message-archiving and replay feature in Amazon SNS FIFO topics?
Open an interactive chat with Bash
Why does using AWS CloudTrail data events for replay not work in this scenario?
Open an interactive chat with Bash
What are the key differences between Amazon SNS FIFO topics and Amazon SQS Standard queues in this use case?
Open an interactive chat with Bash
AWS Certified Solutions Architect Professional SAP-C02
Design for New Solutions
Your Score:
Report Issue
Bash, the Crucial Exams Chat Bot
AI Bot
Loading...
Loading...
Loading...
IT & Cybersecurity Package Join Premium for Full Access