Does pyspark require py4j to be installed (via pip install py4j)?
In 5.5 branch, ran:
mvn clean package -Pspark-1.5 -Dhadoop.version=2.2.0 -Phadoop-2.2 -DskipTests
-P build-distr
Build passed. Ran zeppelin from distribution directory
Did not alter or create zeppelin-env.sh or zeppelin-site.xml (left as templates)
./bin/zeppelin-daemon.sh start
Created a simple notebook
%pyspark
i = 1
pyspark is not responding Traceback (most recent call last):
File "/tmp/zeppelin_pyspark.py", line 20, in <module>
from py4j.java_gateway import java_import, JavaGateway, GatewayClient
ImportError: No module named py4j.java_gateway
On 11/7/15, 1:56 AM, "moon soo Lee" <[email protected]> wrote:
>Thanks for verifying.
>
>Yes that's right.
>pyspark in -all archive will work out of box without any configuration.
>
>Best,
>moon
>
>On Sat, Nov 7, 2015 at 6:49 PM tog <[email protected]> wrote:
>
>> Hi
>>
>> Before casting my vote I was willing to give a try to pyspark (which I have
>> never done so far). Is it supposed to work out of the box in local just
>> untaring the -all archive or do I have to configure something?
>>
>> Cheers
>> Guillaume
>>
>>
>> --
>> PGP KeyID: 2048R/EA31CFC9 subkeys.pgp.net
>>
