Question

A company runs a serverless order-processing system where an EventBridge custom event bus routes 'OrderPlaced' events to a Lambda function that writes to DynamoDB. During a recent 30-minute regional degradation, the Lambda target repeatedly failed and the DLQ was not configured, so an unknown number of order events were lost. Leadership now requires that the team be able to recover and reprocess any events that fail to reach or be successfully processed by targets during future incidents, with no code changes to producers. Which combination of EventBridge features BEST meets this requirement?

Accepted Answer

Enable EventBridge archive on the event bus with an appropriate retention period, and configure an SQS dead-letter queue plus a retry policy on the Lambda target; use archive replay to reprocess events after an incident.

Answer

Increase the Lambda function's reserved concurrency and enable provisioned concurrency so no invocations are dropped during regional degradation.

Answer

Add a second EventBridge rule that forwards all events to an S3 bucket via Kinesis Data Firehose, then write a custom Lambda to parse and replay S3 objects during incidents.

Answer

Enable CloudTrail data events for the event bus and use Athena to query and manually re-emit lost events through the PutEvents API.