What's New in Cloudera Manager 7.4.2

Cloudera Bug: OPSAPS-55802: New Atlas Health Test checks health of the Atlas UI

Previously, Cloudera Manager showed that the health of the Atlas service was good, even when the Atlas UI was not available. A new Health test was added for Atlas which checks the availability of the Atlas UI. This guarantees that Cloudera Manager shows Atlas's health as bad when the Atlas UI is not available. The health test is enabled by default.

Cloudera Bug: OPSAPS-60146ConsumerEmission Timeout for Streams Messaging Manager is now configurable

Consumer metrics emission timeout is configurable for Streams Messaging Manager using the cm.metrics.emit.consumer.metrics.timeout configuration property.

Cloudera Bug: OPSAPS-60223: Change Schema Registry log level to WARN for third-party components

The log messages produced by Ranger are moved to a separate log file, keeping the schema registry log clean.

OPSAPS-59410: BasicAuthentication over TLS support for Kafka Metrics

Added TLS and BasicAuth support to the Kafka Brokers' metrics-related HTTP endpoints. - If both the kafka.http.metrics.ssl.enabled and ssl_enabled configuration properties are enabled, the endpoints will be encrypted via TLS. - If kafka.http.metrics.authentication.enabled is enabled, BasicAuth will be enabled, the username and password will have to be provided in the following properties: kafka.http.metrics.user and kafka.http.metrics.password.

OPSAPS-59446: Kafka client's configuration in Streams Messaging Manager has been simplified

A Kafka clients' security.protocol is auto-configured in Streams Messaging Manager configuration. This change removes the 'Kafka Client Security Protocol' parameter where it was set manually. the Jaas configuration for Kafka Clients can be overridden by using the streams.messaging.manager.kafka.client.sasl.jaas.config property in the Advanced Configuration Snippet for streams-messaging-manager.yaml.

OPSAPS-59530: Service Monitor and Host Monitor scaling improvements

In the CDP Public Cloud environment, the Service Monitor and Host Monitor now adapt their memory settings to improve cluster scaling. This requires the "Automatic Restart Process" to be enabled. (It is enabled by default).