Trying to create phoenix interpreter using %jdbc in Zeppelin using 2.5

[#1]

I am trying to create phoenix interpreter using %jdbc in Zeppelin using 2.5 and am not succeeding.

Steps are:

1. Log into Zeppelin (sandbox 2.5)
2. Create new interpreter as follows

   

3. restart (just to be paranoid)
4. go to my notebook and bind interpreter

   

5. when I run with %jdbc(phoenix) I get Prefix not found.
6. when I run it with %jdbc.phoenix I get jdbc.phoenix interpreter not found

What am I missing?

[#2 · Accepted solution]

Hi, @gkeys, following the RedShift example here, the properties in your JDBC interpreter settings would be phoenix.user, phoenix.driver, phoenix.url, instead. Since you have modified the "default" prefix, does the plain "%jdbc" not work for the interpreter setting?

[#3]

@Randy Gelhausen

Thanks. What threw me off is that when creating a new jdbc interpretter (at least in sandbox) it is prepopulutated with default prefix properties and psql values.. Did not know that entire property and value needed to be deleted and recreated with new prefix (vs only new values).

[#4]

I just installed HDP 2.5 from Ambari on a small openstack cluster choosing Zeppelin from ambari.

%jdbc(phoenix)

select * from phillycrime limit 100

worked for me.

Try to restart the jdbc interpreter

Make sure you have:

under:

jdbc %jdbc (default)

edit restart remove

Option

shared Interpreter for note

Connect to existing process

phoenix.driver org.apache.phoenix.jdbc.PhoenixDriver

phoenix.hbase.client.retries.number 1

phoenix.password

phoenix.url jdbc:phoenix:192.168.1.100:/hbase-unsecure

phoenix.user phoenixuser

[#5]

@Timothy Spann very effective answer, but similar information as @Randy Gelhausen and he was first in.