MapR Integration¶

This document describes details when installing Fire Insights on a MapR cluster.

Download Fire Insights¶

Download MapR specific binary from : https://www.sparkflows.io/archives

Turn off Impersonation¶

In Administration / Configuration of Sparkflows:

Turn off impersonation : Set app.impersonateUsers = false
Set maprfs : hdfs.namenodeURI = maprfs:///
Set spark-submit appropriately : spark.spark-submit = /opt/mapr/spark/xyz/bin/spark-submit

Update http port¶

Set http port` to be different in `conf/application.properties if there are other processes using the specified ports

Fire User¶

Fire has to be installed as a user which can submit jobs to the MapR cluster. Say we installed Fire as user mapr:
```
Create a mapr user in sparkflows and log in as that user
Start using Sparkflows
```