Cloudera Hue

What is Hue

Hue is an open source SQL Assistant for Databases & Data Warehouses.

Preparation

Get Kyuubi Started

Get the server Started first before your try Hue with Kyuubi.

Welcome to
  __  __                           __
 /\ \/\ \                         /\ \      __
 \ \ \/'/'  __  __  __  __  __  __\ \ \____/\_\
  \ \ , <  /\ \/\ \/\ \/\ \/\ \/\ \\ \ '__`\/\ \
   \ \ \\`\\ \ \_\ \ \ \_\ \ \ \_\ \\ \ \L\ \ \ \
    \ \_\ \_\/`____ \ \____/\ \____/ \ \_,__/\ \_\
     \/_/\/_/`/___/> \/___/  \/___/   \/___/  \/_/
                /\___/
                \/__/

Run Hue in Docker

Here we demo running Kyuubi on macOS and Hue on Docker for Mac, there are several known limitations of network, and you can find workarounds from here.

Configuration

1. Copy a configuration template from Hue Docker image.

docker run --rm gethue/hue:latest cat /usr/share/hue/desktop/conf/hue.ini > hue.ini

1. Modify the hue.ini

[beeswax]
  # Kyuubi 1.1.x support thrift version from 1 to 10
  thrift_version=7
  # change to your username to avoid permissions issue for local test
  auth_username=chengpan

[notebook]
  [[interpreters]]
    [[[sql]]]
      name=SparkSQL
      interface=hiveserver2
      
[spark]
  # Host of the Spark Thrift Server
  # For macOS users, use docker.for.mac.host.internal to access host network
  sql_server_host=docker.for.mac.host.internal

  # Port of the Spark Thrift Server
  sql_server_port=10009
  
# other configurations
...

Start Hue in Docker

docker run -p 8888:8888 -v $PWD/hue.ini:/usr/share/hue/desktop/conf/hue.ini gethue/hue:latest

Go http://localhost:8888/ and follow the guide to create an account.

Having fun with Hue and Kyuubi!

For CDH 6.x Users

If you are using CDH 6.x, there is a trick that CDH 6.x blocks Spark in default, you need to modify the configuration to overwrite the desktop.app_blacklist to remove this restriction.

Config Hue in Cloudera Manager.

Refer following configuration and tune it to fit your environment.

[desktop]
 app_blacklist=zookeeper,hbase,impala,search,sqoop,security
 use_new_editor=true
[[interpreters]]
[[[sparksql]]]
  name=Spark SQL
  interface=hiveserver2
  # other interpreters
  ...
[spark]
sql_server_host=kyuubi-server-host
sql_server_port=10009

You need to restart the Hue Service to activate the configuration changes, and then Spark SQL will available in editor list.

Having fun with Hue and Kyuubi!