History

Origin

Strimzi started inside Red Hat in 2017. A small team of three developers looked at how to run stateful workloads like Apache Kafka on Kubernetes (Red Hat Developer). The first public release, 0.1.0, shipped in January 2018 as a set of Docker images and Kubernetes YAML. The 0.2.0 release in March 2018 was the first version that looked like the operator pattern Strimzi uses today (Strimzi incubation blog).

Timeline

Year	Milestone
2017	Red Hat begins exploring Kafka on Kubernetes with three developers
2018	0.1.0 (Docker images plus YAML); 0.2.0 adopts the operator pattern
2019	Donated to the CNCF and accepted as a Sandbox project on 2019-08-28
2024	Promoted to CNCF Incubating on 2024-02-08

How it evolved

The project moved from a bundle of images and manifests to a full operator that reconciles Kafka clusters from custom resources. Strimzi was accepted into the CNCF Sandbox on 2019-08-28 (CNCF projects) and promoted to Incubating on 2024-02-08. At promotion the project reported more than 1600 contributors, more than 180 contributing organisations, and 15 public adopters (CNCF blog).

The largest recent shift follows Apache Kafka itself. The code at this commit has dropped ZooKeeper entirely and runs KRaft only, because Kafka 4.x (4.3.0 here, pom.xml:87) is KRaft only. KRaft metadata handling lives in KRaftMetadataManager.java and KRaftVersionChangeCreator.java. The Kafka custom resource has also reached the v1 API version (api/src/main/java/io/strimzi/api/kafka/model/kafka/Kafka.java:80), promoted from the earlier v1beta2, which lines up with the 1.0 major release series.

Where it stands now

Strimzi is a CNCF Incubating project under neutral governance defined in a separate governance repository. The pinned commit 9505103 (2026-06-23) sits on main after the 1.0.1 stable release (2026-06-17) and around the 1.1.0-rc1 (2026-06-22) line. Releases continue on a regular cadence tracking upstream Apache Kafka versions.