Logging

Learn how you can configure logging for the Strimzi Cluster Operator, Kafka, and Kafka cluster components.

Configuring logging for the Strimzi Cluster Operator

You configure logging for the Strimzi Cluster Operator by editing the strimzi-cluster-operator ConfigMap.

Strimzi Cluster Operator logging is configured through a ConfigMap named strimzi-cluster-operator. This ConfigMap is created when the Strimzi Cluster Operator is installed. It uses log4j2 configuration.

You can configure logging by modifying the data.log4j2.properties values in the ConfigMap. Use kubectl edit to modify the ConfigMap.

kubectl edit configmap strimzi-cluster-operator \
  --namespace [***STRIMZI CLUSTER OPERATOR NAMESPACE***]

The ConfigMap enables you to configure various aspects of logging, including the interval at which logging configuration is reloaded, the root logger level, the log output format, and the log levels for different components. You can also control the logging levels for the Kafka AdminClient, ZooKeeper ZKTrustManager, and Netty. Applying the new logging configuration is dynamic, you do not need to restart the Strimzi Cluster Operator.

Configuring logging for Kafka cluster components

Learn how to configure logging for Kafka cluster components. You can configure logging for these components directly in the Kafka resource, or by referencing a ConfigMap.

The logging properties of Kafka cluster components like Kafka brokers, ZooKeeper, Cruise Control, and all other components deployed and managed through the Kafka resource are configured in the Kafka resource.

Logging properties are specified in spec.[***COMPONENT***].logging. Logging properties can be added directly to this property, or can be defined in an external ConfigMap that is referenced in the Kafka using configMapKeyRef property.

You choose the configuration method by setting the logging.type property to either inline or external.

Inline

Inline configuration means that you directly specify the logging properties in the Kafka resource at the spec of each component.

#...
kind: Kafka
spec:
  #...
    logging:
      type: inline
      loggers:
        kafka.root.logger.level: INFO

External

External configuration means that you reference your own ConfigMap that holds the logging properties.

#...
kind: Kafka
spec:
  #...
    logging:
      type: external
      valueFrom:
        configMapKeyRef:
          name: my-config-map
          key: my-config-map-key

A ConfigMap is generated for each Kafka cluster component after pod creation. These ConfigMaps contain the actual logging configuration. Do not edit the generated ConfigMaps directly, as direct changes are ignored.

Depending on the changes made, they are either applied dynamically, or a rolling restart is triggered.

The following Kafka cluster components use log4j configuration:

Kafka
ZooKeeper

The following Kafka cluster components use log4j2 configuration:

CruiseControl
UserOperator
EntityOperator