What's New in Hue

Learn about the new features of Hue in Cloudera Runtime 7.1.8.

Security and performance improvements by upgrading to Python 3

Python 2 has reached the end of life and is no longer supported. Hue now uses Python 3 which makes use of critical bug fixes and Common Vulnerabilities and Exposures (CVE) fixes for many third-party software dependencies. Python libraries have been upgraded from Python 2.7 to Python 3.8.

Python modules such as django-auth-ldap, django-axes, djangorestframework-simplejwt, Mako, Markdown, python-ldap, django-babel, django-mako, django-cors-headers, djangorestframework, eventlet, sqlparse, and so on have also been upgraded.

These upgrades bring in significant performance improvement and stability in query execution. Operating System, Python version, and Python module upgrades have resulted in a more stable environment.

You must install Python 3.8 after installing Cloudera Manager on your cluster. For more information, see Installing Python 3.8.

You also need to install Psycopg2 package for PostgreSQL database and MySQL client for MySQL and MariaDB databases. For more information, see Installing the psycopg2 Python package for PostgreSQL-backed Hue, Installing the MySQL client for MySQL database, Installing the MySQL client for MariaDB database.

CDP 7.1.8 uses Django version 3.2.13 which supports Oracle Database Server versions 12.2 and higher and cx_Oracle (the Python interface for Oracle database) versions 6.0 or higher.

Hue uses Gunicorn instead of the CherryPy web server

In CDP 7.1.8, Cloudera Manager starts Hue with Gunicorn web server instead of CherryPy.

Hue uses TLS 1.2 by default

On CDP 7.1.8 and higher, Hue and Hue Load Balancer use TLS 1.2 or 1.3 by default. You no longer have to configure settings in Cloudera Manager (7.7.1) to enforce TLS 1.2.

High Availability support added for Oracle databases

Hue supports Oracle Real Application Clusters (RAC) on the Oracle database to achieve high availability and seamlessly handle SELECT, DDL, and DML operations. For more information, see Configuring high availability support for Oracle database.

No 64-character restriction on hostnames for Hue roles

Hue supports creating Hue role hostnames of more than 64 characters. There is no longer a restriction of 64 characters for the “BalancerMember Route” property.

Hue uses the SHA-256 signing algorithm for SAML authentication

SHA-1 signing algorithm is deprecated in most environments. Hue now uses a stronger, secure hash algorithm, SHA-256, for signature and digest methods when authenticating using SAML.

Hue uses basic search to look up data in Atlas

Hue polls Atlas for changes to the entities at a set frequency. By default, Hue now uses the basic search option to look up data in Atlas which is more efficient than the Domain-Specific Language (DSL). DSL queries are resource-intensive and thus expensive. You can enable DSL search by setting the search_cluster property in the Advanced Configuration Snippet. See Enabling DSL search for Hue.

Update to the list of supported non-ASCII characters

Hue supports an additional set of non-ASCII characters. Review the list: Supported non-ASCII and special characters in Hue.

Support for HiveServer2 (HS2) high availability

Hue can now handle HS2 failover using ZooKeeper without setting up a load balancer. You must configure the following setting in the Hue Advanced Configuration snippet:
[beeswax]
hive_discovery_hs2=true
hive_discovery_hiveserver2_znode=/hiveserver2
See Configuring Hue to handle HS2 failover.

Hue supports rolling restart

Hue service downtime is reduced from more than 30 minutes to approximately 80-90 seconds when you restart the CDP cluster in the rolling restart mode. When you restart only the Hue service, then Hue’s non-worker roles, such as the load balancer, Kerberos ticket renewer, and Hue server restart one after the other. For information about the rolling restart options, see Options to restart the Hue service.

Ability to view Hive query details in Hue

Hue packs the combined abilities of Data Analytics Studio (DAS) such as query optimization, query debugging framework, and rich query editor experience of Hue, making Hue the next generation SQL assistant on CDP. You can search Hive query history, view query details, visual explain plan, and DAG information, compare two queries, and download debug bundles for troubleshooting from the Job Browser page.

To support this feature, a new service called Query Processor is added to the CDP stack as a dependency for Hue. It is used for indexing and retrieving Hive query history and query details. You must install the Query Processor service manually on your CDP Private Cloud Base clusters. To know more about the Query Processor service, see About Hue Query Processor. For installation instructions, see Adding Query Processor service to a cluster.

Hue supports Spark SQL using Apache Livy for Spark 3

Hue supports Spark 3 and Livy for Spark 3. To run Spark SQL queries, you must install the Spark 3 and Livy for Spark 3 services on your CDP cluster and then configure these services using Cloudera Manager. See Enabling Spark 3 engine in Hue.

Added Phoenix as a new supported dialect

Apache Phoenix provides a SQL interface to create, access, and query HBase tables using the Hue query editor. You can query HBase tables using the Phoenix SQL Data Definition Language (DDL) and Data Manipulation Language (DML) commands and grammar. For more information, see Enabling the Phoenix SQL editor in Hue.

Hue scripts included in CDP

CDP now includes Hue scripts that you can use for cleaning up old data, setting default editors, changing the document owners, and so on. You no longer need to clone the scripts from Cloudera’s GitHub repository. For more information, see Using Hue scripts.