How to activate hive interpreter for Apache Zeppelin on Dataproc on GCP?

Question

Hive is not enabled by default on GCP on Apache Zeppelin on Dataproc on GCP. How can we activate it?

Answer 1

We can activate Hive interpreter by creating a new interpreter with some changes in the properties and dependencies

1. Set hive.url as
jdbc:hive2://localhost:10000

2. Remove username and password (keep them empty)

3. Set hive.driver property as:
org.apache.hive.jdbc.HiveDriver

4. Adding two artifact under dependencies tab:
org.apache.hive:hive-jdbc:0.14.0
org.apache.hadoop:hadoop-common:2.6.0

You can watch this video for more explanations and code is here

Answer 2

There are multiple ways to use Hive on Dataproc :

either using Hive jobs API

or Spark SQL.

But for activating it, I just  wrote hive as a command in SSH of master node and it started to work. It seems it is already in hadoop enviroment and does not need further activation.

Comments

This is not the requested solution