this QuickStart guide mentions configuring JAVA_HOME in hadoop-env.sh. This is supposed to be located in the hadoop installation directory. However, I installed this from a 1.0.4 .deb package on Ubuntu 12.04. This creates multiple hadoop-env.sh and the startup scripts at /usr/sbin. I'm not sure which hadoop-env.sh i should use for my configuration. A related question is what directory should be specified as HADOOP_HOME.
from the wiki:
HADOOP_HOME is the parent directory of the bin directory that holds the Hadoop scripts
i'm sure they don't mean the parent of /usr/sbin though.
$> locate hadoop-env.sh
/etc/default/hadoop-env.sh /etc/hadoop/hadoop-env.sh /etc/profile.d/hadoop-env.sh /usr/sbin/update-hadoop-env.sh /usr/share/hadoop/templates/conf/hadoop-env.sh
I am also using this for reference since I'm setting up a single-node cluster. I think I downloaded the deb package from Cloudera but I can't recall at the moment and I'm considering restarting the installation