Hadoop : JPS can not find Java installed

Question

my configurations are

hduser@worker1:/usr/local/hadoop/conf$ jps
The program 'jps' can be found in the following packages:
 * openjdk-6-jdk
 * openjdk-7-jdk
Ask your administrator to install one of them

I have java installed though

hduser@worker1:/usr/local/hadoop/conf$ java -version
java version "1.6.0_23"
OpenJDK Runtime Environment (IcedTea6 1.11pre) (6b23~pre10-0ubuntu5)
OpenJDK 64-Bit Server VM (build 20.0-b11, mixed mode)

hduser@worker1:/usr/local/hadoop/conf$ echo $JAVA_HOME 
/usr/lib/jvm/java-1.6.0-openjdk

and also set up in conf/hadoop-env.sh

hduser@worker1:/usr/local/hadoop/conf$ cat hadoop-env.sh | grep JAVA_HOME
# The only required environment variable is JAVA_HOME.  All others are
# set JAVA_HOME in this file, so that it is correctly defined on
 export JAVA_HOME=/usr/lib/jvm/java-1.6.0-openjdk

How can I make JPS work?

Solution 1

It seems like open-jdk does not have jps in it. For hadoop, installing sun-jvm would be a better choice.

Solution 2

Saurabh Saxena's answer above is no longer correct. To get jps, you want to also install the development tools java-1.6.0-openjdk-devel. On CentOS 6 the file is: java-1.6.0-openjdk-devel.x86_64

So:

yum install java-1.6.0-openjdk*

will do the trick (also picks up demo and javadocs besides the jdk and dev tools, but you will get the full complement of command line tools).

For Ubuntu:

apt-get install java-1.6.0-openjdk-devel 

For all these examples, you can try JDK7 (just substitute 1.7), and as of December 2012, Hadoop is pretty stable without the Oracle libraries. See: http://openjdk.java.net/install/

Solution 3

This might also be a reason. Its simple: See if $javac works. Note: $java might work, check javac. If $javac is not working then $jps will not work either. So you might want to do something like

export PATH=$PATH:$JAVA_HOME/bin

and try again. both javac and jps. good luck.

Solution 4

I have found the solution for the missing JPS command. I was installing Hadoop 1.x on ubuntu machine in a pseudo distributed mode. I used Java-7-openJDK to provide for the Java commands and tools. For some reason there was a java-1.6.0-openjdk-devel for version 6 but none for version 7 specifically debian and ubuntu distributions. I am not sure if the same is true for Fedora and Redhat. So the best answer as that time was using the linux command

ps -aux | grep java

I hated doing that because Hadoop daemons start with so many options that each result fills up more than a screen. Apart from seeing that java is running it is impossible to see what hadoop daemons are running. Hence i came up with a short soultion in the form of one line shell script

This is my JPS scirpt for open JDK

!#/bin/bash

ps -aux | grep java | awk '{print $12}'

END

I saved these two lines in a file named jps and stored it in the hadoop/bin directory with execute permissions

**Here is the result of the script hduser@localhsot# ./jps

-Dproc-namenode

-Dproc-datanode

-Dproc-JobTracker

-Dproc-TaskTracker**

Solution 5

try this....

sudo apt-get install openjdk-7-jdk

View more solutions