Migrate from using Debezium Embedded to Debezium Server.

Configure a Quarkus application with production parameters.

Install and manage an Apache Kafka cluster in a Kubernetes cluster.

In this article, we’ll explore how to put everything in production more concretely, in Kubernetes, we’ll learn how to:

In all three previous articles, we’ve been learning all these technologies from the developer's point of view, and we end up deploying the application locally (in the dev mode) on the developer’s machine. Then in part 3, we covered the Outbox pattern and Debezium to avoid the dual-writes problem that occurs when different systems need to synchronize the same data. The following figure shows the architecture of the application: We post-processed these events in real-time to count the number of times a movie is played for more than 10 minutes. We stored Movies in one Kafka topic and, in another Kafka topic, held each occurrence when a user stopped watching a movie and captured the time it had been played. In that example, we simulated a Movies streaming company.

In part 1 and part 2 of this series, we learned about the integration between Apache Kafka, Apache Kafka Streams, and Quarkus, where we developed a simple application producing events to a Kafka topic and consuming and processing them in real-time with Kafka Streams.

Arch virtualbox select label type