Question

A retail company runs an event-driven order platform where dozens of microservices publish and consume events through a central Amazon EventBridge event bus. Recently, several downstream consumer services began failing intermittently because producer teams changed event payload structures without notifying consumers. There is no formal contract between teams, and consumers hardcode field parsing logic. The operations team wants to improve reliability by giving consumer teams a way to discover event structures, generate strongly typed bindings, and detect breaking changes early, without forcing every producer to change how it publishes. Which approach best improves the workload?

Accepted Answer

Enable EventBridge Schema Registry with schema discovery on the event bus, and have consumer teams generate code bindings from the discovered schemas while versioning schemas to catch breaking changes.

Answer

Route all events through an Amazon SQS FIFO queue in front of each consumer so that malformed events are retried until the payload eventually matches.

Answer

Require every producer to publish events to a dedicated Amazon SNS topic and have consumers subscribe with filter policies that validate the full payload structure.

Answer

Add an AWS Lambda transformer on each consumer that inspects incoming events and rewrites unknown fields to default values before processing.