Main Changes since 0.18.0
- Add support for authorization using Open Policy Agent
- Add support for scale subresource to make scaling of following resources easier:
- Remove deprecated
Kafka.spec.topicOperatorclasses and deployment logic
- Use Java 11 as the Java runtime
- Removed the need to manually create Cruise Control metrics topics if topic auto creation is disabled.
- Migration to Helm 3
- Refactored the format of the
KafkaRebalanceresource's status. The state of the rebalance is now displayed in the associated
typefield rather than the
statusfield. This was done so that the information would display correctly in various Kubernetes tools.
- Added performance tuning options to the
KafkaRebalanceCR and the ability to define a regular expression that will exclude matching topics from a rebalance optimization proposal.
- Use Strimzi Kafka Bridge 0.18.0
- Make it possible to configure labels and annotations for secrets created by the User Operator
- Strimzi Kafka Bridge metrics integration:
- enable/disable metrics in the KafkaBridge custom resource
- new Grafana dashboard for the bridge metrics
- Support dynamically changeable logging in the Entity Operator and Kafka Bridge
Full list of changes can be found under the 0.19.0 milestone.
Deprecations and removals
Deprecation of Helm v2 chart
The Helm v2 support will end soon.
Bug fixing should stop on August 13th 2020 and security fixes on November 13th.
See https://helm.sh/blog/covid-19-extending-helm-v2-bug-fixes/ for more details.
In sync with that, the Helm v2 chart of Strimzi Cluster Operator is now deprecated and will be removed in the future as Helm v2 support ends.
Since Strimzi 0.19.0, we have a new chart for Helm v3 which can be used instead.
Removal of v1alpha1 versions of several custom resources
In Strimzi 0.12.0, the
v1alpha1 versions of the following resources have been deprecated and replaced by
In the next release, the
v1alpha1 versions of these resources will be removed.
Please follow the guide for upgrading the resources: https://strimzi.io/docs/operators/latest/deploying.html#assembly-upgrade-str.
Removal deprecated cadvisor metric labels
container_name labels provided on the cadvisor metrics are now just
container starting from Kubernetes 1.16.
We removed the old ones from the Prometheus scraping configuration/alerts and on the Kafka and ZooKeeper dashboard as well.
It means that the charts related to memory and CPU usage are not going to work on Kuvbernetes version previous 1.14.
For more information on what is changed: #3312
Deprecation of monitoring port on Kafka and ZooKeeper related services
PodMonitor resource is now used instead of the
ServiceMonitor for scraping metrics from Kafka, ZooKeeper, Kafka Connect and so on.
For this reason, we are deprecating the monitoring port
tcp-prometheus (9404) on all the services where it is declared (Kafka bootstrap, ZooKeeper client and so on).
This port will be removed in the next release.
Together with it we will also remove the Prometheus annotation from the service.
Removal warning of Cluster Operator log level
Because of the new Cluster Operator dynamic logging configuration via PR#3328 we are going to remove the
STRIMZI_LOG_LEVEL environment variable from the Cluster Operator deployment YAML file in the 0.20.0 release.