chore(deps): update quay.io/strimzi/operator docker tag to v0.37.0
This MR contains the following updates:
Package | Type | Update | Change |
---|---|---|---|
quay.io/strimzi/operator | ironbank-docker | minor |
0.36.1 -> 0.37.0
|
quay.io/strimzi/operator | minor |
0.36.1 -> 0.37.0
|
|
quay.io/strimzi/operator | stage | minor |
0.36.1 -> 0.37.0
|
⚠ Dependency Lookup Warnings ⚠
Warnings were logged while processing this repo. Please check the logs for more information.
Release Notes
strimzi/strimzi-kafka-operator
v0.37.0
- The
StableConnectIdentites
feature gate moves to beta stage. By default, StrimziPodSets are used for Kafka Connect and Kafka Mirror Maker 2. If needed,StableConnectIdentites
can be disabled in the feature gates configuration in the Cluster Operator. - Support for the
ppc64le
platform - Added version fields to the
Kafka
custom resource status to track install and upgrade state - Support for infinite auto-restarts of Kafka Connect and Kafka Mirror Maker 2 connectors
Changes, deprecations and removals
-
Removed support for OpenTracing:
- The
tracing.type: jaeger
configuration, inKafkaConnect
,KafkaMirrorMaker
,KafkaMirrorMaker2
andKafkaBridge
resources, is not supported anymore. - The OpenTelemetry based tracing is the only available by using
tracing.type: opentelemetry
.
- The
-
The default behavior of the Kafka Connect connector auto-restart has changed.
When the auto-restart feature is enabled in
KafkaConnector
orKafkaMirrorMaker2
custom resources, it will now continue to restart the connectors indefinitely rather than stopping after 7 restarts, as previously. If you want to use the original behaviour, use the.spec.autoRestart.maxRestarts
option to configure the maximum number of restarts. For example:apiVersion: kafka.strimzi.io/v1beta2 kind: KafkaConnector metadata: labels: strimzi.io/cluster: my-connect name: echo-sink-connector spec:
...
autoRestart:
enabled: true
maxRestarts: 7
...
* **The automatic configuration of Cruise Control CPU capacity has been changed in this release**:
* There are three ways to configure Cruise Control CPU capacity values:
* `.spec.cruiseControl.brokerCapacity` (for all brokers)
* `.spec.cruiseControl.brokerCapacity.overrides` (per broker)
* Kafka resource requests and limits (for all brokers).
* The precedence of which Cruise Control CPU capacity configuration is used has been changed.
* In previous Strimzi versions, the Kafka resource limit (if set) took precedence, regardless if any other CPU configurations were set.
* For example:
* (1) Kafka resource limits
* (2) `.spec.cruiseControl.brokerCapacity.overrides`
* (3) `.spec.cruiseControl.brokerCapacity`
* This previous behavior was identified as a bug and was fixed in this Strimzi release.
* Going forward, the brokerCapacity overrides per broker take top precedence, then general brokerCapacity configuration, and then the Kafka resource requests, then the Kafka resource limits.
* For example:
* (1) `.spec.cruiseControl.brokerCapacity.overrides`
* (2) `.spec.cruiseControl.brokerCapacity`
* (3) Kafka resource requests
* (4) Kafka resource limits
* When none of Cruise Control CPU capacity configurations mentioned above are configured, CPU capacity will be set to `1`.
as any _override_ value configured in the `.spec.cruiseControl` section of the `Kafka` custom resource.
Configuration
-
If you want to rebase/retry this MR, check this box
This MR has been generated by Renovate Bot.