Geek Answers Handbook

Apache zeppelin process is dead

I am trying to run zeppelin on Ubuntu14 w / Hadoop 1.0.3 and Spark 1.4.0. I finished creating the source code and the whole package successfully completed the construction. But when I start the demon, it fails and says that the Zeppelin process is dead.

Any ideas where this is going wrong?

It says that he cannot find the logs folder and launcher folder that definitely exist.

+6
source share
6 answers

Joseph

I suggest you first try testing your zeppelin package.

mvn verify

or check if your zeppelin process is alive or not.

 ps -aux | grep zeppelin
+4
source

Try running zeppelin on sudo :

 sudo bin/zeppelin-daemon.sh start
+2
source

It may be a bug caused by the JDK version, at least it was for me try updating jdk and creating it again.

Also, make sure you create it using the correct command

 mvn clean package -Pspark-1.4 -Dhadoop.version=2.2.0 -Phadoop-2.2 -DskipTests
+1
source

This works for me:

 ps -ef | grep "zeppelin" kill -9 pid sudo bin/zeppelin-daemon.sh restart
+1
source

I had the same problem and tried the suggested answers, but none helped me. Here is what worked for me:

  • Download the binaries , and then download the build requirements:

     sudo apt install openjdk-8-jdk npm libfontconfig r-base-dev r-cran-evaluate sudo apt install maven

  • Go to the Zeppelin directory and run:

     sudo bin/zeppelin-daemon.sh start
  • Go to localhost:8080 in your browser.
0
source

If you are running zeppelin from a virtual machine, make sure you have enough RAM and CPU. I ran into your problem when I used Virtual Box and the default settings. When I increased the processor to 2 and RAM to 4096, everything worked fine. This is because zeppelin starts the spark by default, and the spark is very resource-intensive locally and otherwise.

-1
source

More articles:


All Articles