Confirmed, building with both -Pr and -Psparkr will causes errors in zeppelin as this is not an expected build combination.
Try building with just one interpreter profile, such as: mvn clean package -DskipTests -Pspark-1.6 -Phadoop-2.6 -Pyarn -Ppyspark -Psparkr Confirmed that R still works on master branch as of today May 8th. Ethan - also be sure you have R and the R dependencies installed. Jeff On 5/7/16, 2:20 AM, "Amos B. Elberg" <amos.elb...@gmail.com> wrote: >Ethan - can you clarify what you mean when you say you couldn't find it? What >specifically fails to happen? Are there any log messages? > >You're attempting to build two separate interpreters, -Pr and -Psparkr both of >which want to bind to the same interpreter name, and I don't believe the >behavior in that case is defined. > >> On May 6, 2016, at 10:15 PM, Ethan Xu <ethan.yifa...@gmail.com> wrote: >> >> Hello, >> >> I couldn't find SparkR interpreter in the default build off the Master >> branch (May 6th). I'm using Zeppelin in local environment with hadoop 2.6 >> and Spark 1.6.2. >> >> I confirmed that SparkR was functioning itself by running >> $spark.path/bin/sparkR. >> >> Here is how I built and started zeppelin: >> >> mvn clean package -DskipTests -Pspark-1.6 -Phadoop-2.6 -Pyarn >> -Ppyspark -Pr -Psparkr >> >> ./bin/zeppelin-daemon.sh start >> >> Did I miss anything? Any help is greatly appreciated. >> >> Best, >> >> Ethan
