[desktop]
app_blacklist=
[notebook]
show_notebooks=true
[spark]
server_url=
http://quickstart.cloudera:8998 # Configure Livy to start in local 'process' mode, or 'yarn' workers.
livy_server_session_kind = yarn
# Additional Settings for Livy server, Server url might be redundant
livy_impersonation_enabled=false
livy_server_session_timeout = 3600000
languages='[{"name": "Scala", "type": "scala"},{"name": "Python", "type": "python"},{"name": "Impala SQL", "type": "impala"},{"name": "Hive SQL", "type": "hive"},{"name": "Text", "type": "text"}]'
Livy was installed with the below environment variables set
export JAVA_HOME=/usr/java/jdk1.7.0_67-clouderaexport HUE_SECRET_KEY=cloudera
export HADOOP_CONF_DIR=/etc/hadoop/conf
export SPARK_HOME=/usr/lib/spark
This works ok for a cluster that does not have Kerberos enabled but I get errors when trying to do this same in a cluster that has Kerberos. I have tried various things but cannot find any pointers online that seem to work. Is anyone able to help by letting me know exactly what I would need to change in Livy and in Hue to make Hue Notebook work when the cluster has Kerberos enabled?