As a Kessel Service Owner, I want a dependable way to capture Zookies in inventory API from Relations/SpiceDB with low overhead in both operational complexity and code maintenance. For service providers not requiring read-after-write, Kafka is our best solution for ensuring updates are pushed to inventory API but we have a few options to weigh on how to implement.

1. Leverage the Consumer client library within Inventory itself (RHCLOUD-38504)
2. Leverage the Debezium Sink for JDBC to write directly to Inventory API's DB

This spike is to uncover the overhead of leveraging the JDBC sink and how we can leverage it, to helps us chose which is the better solution

Done Criteria

• Capture requirements, implementation details, and any potential issues or concerns with leveraging the JDBC Sink for Zookie replication