HUE 3.10 Adding Jars to classpath for Hive Queries

[Chriss Vincelette]

Really enjoying HUE so far with it's ldap, hive, db, and zookeeper integration.

We use EMR almost exclusively so we setup a stand alone HUE console on an ec2-instance and point it to the desired EMR cluster when it's available. (I know EMR will install hue for you, but not fully integrated with all of the stuff I mentioned above).

I'm having some trouble querying some of our tables though. Specifically ones that use CSVSerde.

Whenever we are in the hive query editor and select a table that uses CSVSerde we get the following error message:

Error while processing statement: FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.DDLTask. java.lang.ClassNotFoundException Class com.bizo.hive.serde.csv.CSVSerde not found

I have downloaded the proper jar file but everything I've read so far assumes HUE is installed on one of the nodes (usually the master) of the Hadoop cluster.

How do I add this to the classpath of a standalone HUE?

[Romain Rigaux]

Could you try this? http://gethue.com/hadoop-tutorial-hive-udf-in-1-minute/

--
You received this message because you are subscribed to the Google Groups "Hue-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hue-user+u...@cloudera.org.

[Chriss Vincelette]

Thanks, that worked.

We have about ~30 users of our HUE console with accounts sync'ing via ldap.

Will every user need to do this?

Will they need to repeat it every time the EMR cluster is brought up fresh? Currently we retire the cluster in the evenings and only run it during business hours.

Ideally I'd like to bake this into the HUE configuration so they don't need to worry about it. :)

[Romain Rigaux]

There is probably a way to do it at the HS2/Hive level, e.g. https://issues.apache.org/jira/browse/HIVE-5160 or equivalent

[Chriss Vincelette]

I figured out what I needed to do.

I'm documenting it here for the benefit of others new to HUE like me.

We are using Amazon EMR for our Hadoop clusters.

Beeswax is a total red herring. A while back HUE was tweaked under the covers to actually use hiveserver2 but the references to beeswax remain. Hiveserver2 runs completely on the Hadoop Clusters master and slave nodes.

So in order to make the jars available to every HUE Hive user out of the box what I had to do was add the jars into:

/usr/lib/hive/lib/

On all the master and slave nodes of my EMR cluster. I put the jars in an S3 bucket and wrote a simple bootstrap action to copy the files from s3 into /usr/lib/hive/lib/ when the EMR cluster comes up.

After that my table browsing and querying worked without issue in HUE. :)

Woo hue! :)