Cloudera Docs

Release Notes

Learn about the new features, improvements, known and fixed issues, limitations, and unsupported features in this release of CSA Operator.

What's new

Learn about the new features and notable changes in this release of CSA Operator.

CSA Operator 1.1.1

To learn more about CSA Operator and its typical deployment architecture, see Overview and for the installation instructions, see Installation.

This release of the CSA Operator is based on the Apache Flink Kubernetes Operator (Flink Operator) 1.9.0 and Apache Flink (Flink) 1.19.1. For more information about the supported versions, see the following upstream resources:This version of CSA Operator also includes a Technical Preview of the SQL Stream Builder (SSB). SSB is a comprehensive interactive user interface for creating stateful stream processing jobs using SQL. For more information about SSB and its features, see the What is SQL Stream Builder page.

Flink

Rebase to Flink 1.19.1

Apache Flink 1.19.1 is supported in CSA Operator 1.1.0.

For more information on what is included in the Apache Flink 1.19.1 version, see the Apache Flink 1.19.1 Release Announcement and Release Notes.

Rebase to Flink Operator 1.9.0

Apache Flink Operator 1.9.0 is supported in CSA Operator 1.1.0.

For more information on what is included in the Apache Flink Operator 1.9.0 version, see the Apache Flink 1.19.1 Release Announcement and Release Notes.

SQL Stream Builder

Python UDFs in SQL Stream Builder

This feature allows customers to start using Python for creating User-Defined Functions (UDFs). Cloudera recommends that customers start using Python UDFs for all new developments, and start migrating their JavaScript UDFs to Python to prepare for future upgrades, as Javascript UDFs will be removed in the future due to the deprecation of the Nashorn engine used in JDK 8 and 11.

For more information on using Python UDFs, see Python UDFs.

Global logging configuration for SQL Stream Builder jobs

A new global settings view has been enabled, which currently includes log4j configuration of Flink jobs started in SQL Stream Builder. Users with administrator rights in SQL Stream Builder can set a default logging configuration applied to all SSB jobs , which can be overridden at the job level.

For more information see Adjusting logging configuration in Advanced Settings.

Customizable default Kafka TrustStore configuration in Streaming SQL Console

Customizing default Kafka TrustStore configurations was added to Streaming SQL Console. Kafka TrustStore can be configured during adding Kafka as a Data Source on the UI.

Expose session cluster state and ability to stop it

SQL Stream Builder now displays information about existing session clusters and allows for those clusters to be terminated from the UI. For more information, refer to TO-DO.

Fixed issues

Learn about the fixed issues in this release of CSA Operator.

CSA-5065: Artifact Storage request thread does not timeout when storage is offline, hanging the UI
CSA-5291: [k8s][flink] The default Flink image version should point to the Cloudera repository
CSA-5306: SSB API does not validate catalog type
CSA-5356: Flink docker image permission issue
CSA-5424: Job Settings Logging tab bug

Improvements

CSA-5145: [ssb] Only show connectors and formats which are available on the classpath
CSA-5278: [k8s][flink] Specify default USER in the Flink Dockerfile
CSA-5321 [ssb-ui] Hide Flink Dashboard from UI
CSA-5422: Use cloudnative-pg image for postgres in the operator
CSA-5430: Update the Dockerfile in the flink-tutorials branch of the CSA Operator to use our own image

Known issues

Learn about the known issues in this release of CSA Operator.

Stuck session jobs in Cloudera Streaming Analytics Operator

Session jobs stop running if the session cluster's Job Manager is restarted without High Availability configured. However, because of a Flink bug, such stopped jobs get stuck in RECONCILING/STABLE state and cannot be restarted or deleted.

In such cases, the following is seen when using the kubectl get FlinkSessionJobs -n flink command:

kubectl get FlinkSessionJobs -n flink
NAME                         JOB STATUS    LIFECYCLE STATE
ssb-ssbdefault-testjobname   RECONCILING   STABLE

Delete the jobs first, and then remove the finalizers from the object:

kubectl delete FlinkSessionJob [*** JOB NAME ***] -n flink
kubectl patch FlinkSessionJob [*** JOB NAME ***] -n flink --type=merge -p '{"metadata":{"finalizers":[]}}'
FLINK-33536: S3 filesystem sink and CSV format throws error
When using the Flink Table API CSV streaming sink with the S3 filesystem, the operation fails with IOException: Stream closed.

Unsupported features

Learn what features are unsupported in this release of CSA Operator.

The following upgrade scenarios are not supported in this version of CSA Operator:
  • Upgrading from the Technical Preview version to the Generally Available version
  • Upgrading from the Apache Flink Kubernetes Operator to CSA Operator
This version of the CSA Operator also includes the Technical Preview of SQL Stream Builder (SSB). The following SSB features are not included in this version of CSA Operator:
  • Materialized Views
  • Notifications
  • Sampling
  • Webhook sinks
  • Kafka data transformations