CDH4 installation along with MRv1 from tarball

0
votes

Need some guidance on CDH4 installation from tarballs

I have downloaded two files from " https://ccp.cloudera.com/display/SUPPORT/CDH4+Downloadable+Tarballs "

1) hadoop-0.20-mapreduce-0.20.2+1341 (has only MRv1) 2) hadoop-2.0.0+922 (has HDFS+ Yarn)

I was able to install MRv1 from first file "1) hadoop-0.20-mapreduce-0.20.2+1341 " I have edited the hadoop-env.sh, core-site.xml,hdfs-site.xml and mapred-site.xml present in the conf directory of the untared file. My JobTracker and TaskTracker are working fine.

I could see few new variables to in hadoop-env.sh they are: HADOOP_NAMENODE_OPTS HADOOP_SECONDARYNAMENODE_OPTS HADOOP_DATANODE_OPTS HADOOP_BALANCER_OPTS HADOOP_JOBTRACKER_OPTS

What are these ? Am I supposed to set these?

After untaring "2) hadoop-2.0.0+922" . I can see one more set of config files at /etc/hadoop/ Do I need to use these config files?

I am unable to start NAMENODE and DATANODE services from this. What are the required config changes ? Can any one guide me to the proper link which has all installation steps ? Please Help.

1
hadoop

1 Answers

0
votes

Not sure if its too late to answer but

You need to first go to "hadoop-2.0.0-mr1-cdh4.2.0" folder on your machine then run below command

command to start datanode & namenode are :

for x in /"path to hadoop dir on your machine"/hadoop-2.0.0-mr1-cdh4.2.0/bin/hadoop-* ; do $x start datanode ; done ;

for x in /"path to hadoop dir on your machine"/hadoop-2.0.0-mr1-cdh4.2.0/bin/hadoop-* ; do $x start namenode ; done ;