hadoop2安装脚本

#!/bin/bash
#
# Install Hadoop 2 using pdsh/pdcp where possible.
#
# Command can be interactive or file-based.  This script sets up
# a Hadoop 2 cluster with basic configuration.  Modify data, log, and pid
# directories as desired.  Further configure your cluster with ./conf-hadoop2.sh
#
#

# Basic environment variables.  Edit as necessary
HADOOP_VERSION=2.0.5-alpha
HADOOP_HOME="/opt/hadoop-${HADOOP_VERSION}"
NN_DATA_DIR=/var/data/hadoop/hdfs/nn
SNN_DATA_DIR=/var/data/hadoop/hdfs/snn
DN_DATA_DIR=/var/data/hadoop/hdfs/dn
YARN_LOG_DIR=/var/log/hadoop/yarn
HADOOP_LOG_DIR=/var/log/hadoop/hdfs
HADOOP_MAPRED_LOG_DIR=/var/log/hadoop/mapred
YARN_PID_DIR=/var/run/hadoop/yarn
HADOOP_PID_DIR=/var/run/hadoop/hdfs
HADOOP_MAPRED_PID_DIR=/var/run/hadoop/mapred
HTTP_STATIC_USER=hdfs
YARN_PROXY_PORT=8081

source hadoop-xml-conf.sh
CMD_OPTIONS=$(getopt -n "$0"  -o hif --long "help,interactive,file"  -- "$@")

# Take care of bad options in the command
if [ $? -ne 0 ];
then
  exit 1
fi
eval set -- "$CMD_OPTIONS"

all_hosts="all_hosts"
nn_host="nn_host"
snn_host="snn_host"
dn_hosts="dn_hosts"
rm_host="rm_host"
nm_hosts="nm_hosts"
mr_history_host="mr_history_host"
yarn_proxy_host="yarn_proxy_host"

install()
{
    echo "Copying Hadoop $HADOOP_VERSION to all hosts..."
    pdcp -w ^all_hosts hadoop-"$HADOOP_VERSION".tar.gz /opt

    echo "Copying JDK 1.6.0_31 to all hosts..."
    pdcp -w ^all_hosts jdk-6u31-linux-x64-rpm.bin /opt

    echo "Installing JDK 1.6.0_31 on all hosts..."
    pdsh -w ^all_hosts chmod a+x /opt/jdk-6u31-linux-x64-rpm.bin
    pdsh -w ^all_hosts /opt/jdk-6u31-linux-x64-rpm.bin -noregister 1>&- 2>&-

    echo "Setting JAVA_HOME and HADOOP_HOME environment variables on all hosts..."
    pdsh -w ^all_hosts 'echo export JAVA_HOME=/usr/java/jdk1.6.0_31 > /etc/profile.d/java.sh'
    pdsh -w ^all_hosts "source /etc/profile.d/java.sh"
    pdsh -w ^all_hosts "echo export HADOOP_HOME=$HADOOP_HOME > /etc/profile.d/hadoop.sh"
    pdsh -w ^all_hosts 'echo export HADOOP_PREFIX=$HADOOP_HOME >> /etc/profile.d/hadoop.sh'
    pdsh -w ^all_hosts "source /etc/profile.d/hadoop.sh"

    echo "Extracting Hadoop $HADOOP_VERSION distribution on all hosts..."
    pdsh -w ^all_hosts tar -zxf /opt/hadoop-"$HADOOP_VERSION".tar.gz -C /opt

    echo "Creating system accounts and groups on all hosts..."
    pdsh -w ^all_hosts groupadd hadoop
    pdsh -w ^all_hosts useradd -g hadoop yarn
    pdsh -w ^all_hosts useradd -g hadoop hdfs
    pdsh -w ^all_hosts useradd -g hadoop mapred

    echo "Creating HDFS data directories on NameNode host, Secondary NameNode host, and DataNode hosts..."
    pdsh -w ^nn_host "mkdir -p $NN_DATA_DIR && chown hdfs:hadoop $NN_DATA_DIR"
    pdsh -w ^snn_host "mkdir -p $SNN_DATA_DIR && chown hdfs:hadoop $SNN_DATA_DIR"
    pdsh -w ^dn_hosts "mkdir -p $DN_DATA_DIR && chown hdfs:hadoop $DN_DATA_DIR"

    echo "Creating log directories on all hosts..."
    pdsh -w ^all_hosts "mkdir -p $YARN_LOG_DIR && chown yarn:hadoop $YARN_LOG_DIR"
    pdsh -w ^all_hosts "mkdir -p $HADOOP_LOG_DIR && chown hdfs:hadoop $HADOOP_LOG_DIR"
    pdsh -w ^all_hosts "mkdir -p $HADOOP_MAPRED_LOG_DIR && chown mapred:hadoop $HADOOP_MAPRED_LOG_DIR"

    echo "Creating pid directories on all hosts..."
    pdsh -w ^all_hosts "mkdir -p $YARN_PID_DIR && chown yarn:hadoop $YARN_PID_DIR"
    pdsh -w ^all_hosts "mkdir -p $HADOOP_PID_DIR && chown hdfs:hadoop $HADOOP_PID_DIR"
    pdsh -w ^all_hosts "mkdir -p $HADOOP_MAPRED_PID_DIR && chown mapred:hadoop $HADOOP_MAPRED_PID_DIR"

    echo "Editing Hadoop environment scripts for log directories on all hosts..."
    pdsh -w ^all_hosts echo "export HADOOP_LOG_DIR=$HADOOP_LOG_DIR >> $HADOOP_HOME/etc/hadoop/hadoop-env.sh"
    pdsh -w ^all_hosts echo "export YARN_LOG_DIR=$YARN_LOG_DIR >> $HADOOP_HOME/etc/hadoop/yarn-env.sh"
    pdsh -w ^all_hosts echo "export HADOOP_MAPRED_LOG_DIR=$HADOOP_MAPRED_LOG_DIR >> $HADOOP_HOME/etc/hadoop/mapred-env.sh"

    echo "Editing Hadoop environment scripts for pid directories on all hosts..."
    pdsh -w ^all_hosts echo "export HADOOP_PID_DIR=$HADOOP_PID_DIR >> $HADOOP_HOME/etc/hadoop/hadoop-env.sh"
    pdsh -w ^all_hosts echo "export YARN_PID_DIR=$YARN_PID_DIR >> $HADOOP_HOME/etc/hadoop/yarn-env.sh"
    pdsh -w ^all_hosts echo "export HADOOP_MAPRED_PID_DIR=$HADOOP_MAPRED_PID_DIR >> $HADOOP_HOME/etc/hadoop/mapred-env.sh"

    echo "Creating base Hadoop XML config files..."
    create_config --file core-site.xml
    put_config --file core-site.xml --property fs.default.name --value "hdfs://$nn:9000"
    put_config --file core-site.xml --property hadoop.http.staticuser.user --value "$HTTP_STATIC_USER"

    create_config --file hdfs-site.xml
    put_config --file hdfs-site.xml --property dfs.namenode.name.dir --value "$NN_DATA_DIR"
    put_config --file hdfs-site.xml --property fs.checkpoint.dir --value "$SNN_DATA_DIR"
    put_config --file hdfs-site.xml --property fs.checkpoint.edits.dir --value "$SNN_DATA_DIR"
    put_config --file hdfs-site.xml --property dfs.datanode.data.dir --value "$DN_DATA_DIR"
    put_config --file hdfs-site.xml --property dfs.namenode.http-address --value "$nn:50070"
    put_config --file hdfs-site.xml --property dfs.namenode.secondary.http-address --value "$snn:50090"

    create_config --file mapred-site.xml
    put_config --file mapred-site.xml --property mapreduce.framework.name --value yarn
    put_config --file mapred-site.xml --property mapreduce.jobhistory.address --value "$mr_hist:10020"
    put_config --file mapred-site.xml --property mapreduce.jobhistory.webapp.address --value "$mr_hist:19888"
    put_config --file mapred-site.xml --property yarn.app.mapreduce.am.staging-dir --value /mapred

    create_config --file yarn-site.xml
    put_config --file yarn-site.xml --property yarn.nodemanager.aux-services --value mapreduce.shuffle
    put_config --file yarn-site.xml --property yarn.nodemanager.aux-services.mapreduce.shuffle.class --value org.apache.hadoop.mapred.ShuffleHandler
    put_config --file yarn-site.xml --property yarn.web-proxy.address --value "$yarn_proxy:$YARN_PROXY_PORT"
    put_config --file yarn-site.xml --property yarn.resourcemanager.scheduler.address --value "$rmgr:8030"
    put_config --file yarn-site.xml --property yarn.resourcemanager.resource-tracker.address --value "$rmgr:8031"
    put_config --file yarn-site.xml --property yarn.resourcemanager.address --value "$rmgr:8032"
    put_config --file yarn-site.xml --property yarn.resourcemanager.admin.address --value "$rmgr:8033"
    put_config --file yarn-site.xml --property yarn.resourcemanager.webapp.address --value "$rmgr:8088"

    echo "Copying base Hadoop XML config files to all hosts..."
    pdcp -w ^all_hosts core-site.xml hdfs-site.xml mapred-site.xml yarn-site.xml $HADOOP_HOME/etc/hadoop/

    echo "Creating configuration, command, and script links on all hosts..."
    pdsh -w ^all_hosts "ln -s $HADOOP_HOME/etc/hadoop /etc/hadoop"
    pdsh -w ^all_hosts "ln -s $HADOOP_HOME/bin/* /usr/bin"
    pdsh -w ^all_hosts "ln -s $HADOOP_HOME/libexec/* /usr/libexec"

    echo "Formatting the NameNode..."
    pdsh -w ^nn_host "su - hdfs -c '$HADOOP_HOME/bin/hdfs namenode -format'"

    echo "Copying startup scripts to all hosts..."
    pdcp -w ^nn_host hadoop-namenode /etc/init.d/
    pdcp -w ^snn_host hadoop-secondarynamenode /etc/init.d/
    pdcp -w ^dn_hosts hadoop-datanode /etc/init.d/
    pdcp -w ^rm_host hadoop-resourcemanager /etc/init.d/
    pdcp -w ^nm_hosts hadoop-nodemanager /etc/init.d/
    pdcp -w ^mr_history_host hadoop-historyserver /etc/init.d/
    pdcp -w ^yarn_proxy_host hadoop-proxyserver /etc/init.d/

    echo "Starting Hadoop $HADOOP_VERSION services on all hosts..."
    pdsh -w ^nn_host "chmod 755 /etc/init.d/hadoop-namenode && chkconfig hadoop-namenode on && service hadoop-namenode start"
    pdsh -w ^snn_host "chmod 755 /etc/init.d/hadoop-secondarynamenode && chkconfig hadoop-secondarynamenode on && service hadoop-secondarynamenode start"
    pdsh -w ^dn_hosts "chmod 755 /etc/init.d/hadoop-datanode && chkconfig hadoop-datanode on && service hadoop-datanode start"
    pdsh -w ^rm_host "chmod 755 /etc/init.d/hadoop-resourcemanager && chkconfig hadoop-resourcemanager on && service hadoop-resourcemanager start"
    pdsh -w ^nm_hosts "chmod 755 /etc/init.d/hadoop-nodemanager && chkconfig hadoop-nodemanager on && service hadoop-nodemanager start"

    pdsh -w ^yarn_proxy_host "chmod 755 /etc/init.d/hadoop-proxyserver && chkconfig hadoop-proxyserver on && service hadoop-proxyserver start"

    echo "Creating MapReduce Job History directories..."
    su - hdfs -c "hadoop fs -mkdir -p /mapred/history/done_intermediate"
    su - hdfs -c "hadoop fs -chown -R mapred:hadoop /mapred"
    su - hdfs -c "hadoop fs -chmod -R g+rwx /mapred"

    pdsh -w ^mr_history_host "chmod 755 /etc/init.d/hadoop-historyserver && chkconfig hadoop-historyserver on && service hadoop-historyserver start"

    echo "Running YARN smoke test..."
    pdsh -w ^all_hosts "usermod -a -G hadoop $(whoami)"
    su - hdfs -c "hadoop fs -mkdir -p /user/$(whoami)"
    su - hdfs -c "hadoop fs -chown $(whoami):$(whoami) /user/$(whoami)"
    source /etc/profile.d/java.sh
    source /etc/profile.d/hadoop.sh
    source /etc/hadoop/hadoop-env.sh
    source /etc/hadoop/yarn-env.sh
    hadoop jar $HADOOP_HOME/share/hadoop/mapreduce/hadoop-mapreduce-examples-$HADOOP_VERSION.jar pi -Dmapreduce.clientfactory.class.name=org.apache.hadoop.mapred.YarnClientFactory -libjars $HADOOP_HOME/share/hadoop/mapreduce/hadoop-mapreduce-client-jobclient-$HADOOP_VERSION.jar 16 10000
}

interactive()
{
    echo -n "Enter NameNode hostname: "
    read nn
    echo -n "Enter Secondary NameNode hostname: "
    read snn
    echo -n "Enter ResourceManager hostname: "
    read rmgr
    echo -n "Enter Job History Server hostname: "
    read mr_hist
    echo -n "Enter YARN Proxy hostname: "
    read yarn_proxy
    echo -n "Enter DataNode hostnames (comma separated or hostlist syntax): "
    read DNS
    echo -n "Enter NodeManager hostnames (comma separated or hostlist syntax): "
    read nms
    
    echo "$nn" > "$nn_host"
    echo "$snn" > "$snn_host"
    echo "$rmgr" > "$rm_host"
    echo "$mr_hist" > "$mr_history_host"
    echo "$yarn_proxy" > "$yarn_proxy_host"
    dn_hosts_var=$(sed 's/\,/\n/g' <<< $dns)
    nm_hosts_var=$(sed 's/\,/\n/g' <<< $nms)
    echo "$dn_hosts_var" > "$dn_hosts"
    echo "$nm_hosts_var" > "$nm_hosts"
    echo "$(echo "$nn $snn $rmgr $mr_hist $yarn_proxy $dn_hosts_var $nm_hosts_var" | tr ' ' '\n' | sort -u)" > "$all_hosts"
}

file()
{
    nn=$(cat nn_host)
    snn=$(cat snn_host)
    rmgr=$(cat rm_host)
    mr_hist=$(cat mr_history_host)
    yarn_proxy=$(cat yarn_proxy_host)
    dns=$(cat dn_hosts)
    nms=$(cat nm_hosts)
    
    echo "$(echo "$nn $snn $rmgr $mr_hist $dns $nms" | tr ' ' '\n' | sort -u)" > "$all_hosts"
}

help()
{
cat << EOF
install-hadoop2.sh
 
This script installs Hadoop 2 with basic data, log, and pid directories.
 
USAGE:  install-hadoop2.sh [options]
 
OPTIONS:
   -i, --interactive      Prompt for fully qualified domain names (FQDN) of the NameNode,
                          Secondary NameNode, DataNodes, ResourceManager, NodeManagers,
                          MapReduce Job History Server, and YARN Proxy server.  Values
                          entered are stored in files in the same directory as this command.
                          
   -f, --file             Use files with fully qualified domain names (FQDN), new-line
                          separated.  Place files in the same directory as this script.
                          Services and file name are as follows:
                          NameNode = nn_host
                          Secondary NameNode = snn_host
                          DataNodes = dn_hosts
                          ResourceManager = rm_host
                          NodeManagers = nm_hosts
                          MapReduce Job History Server = mr_history_host
                          YARN Proxy Server = yarn_proxy_host
                          
   -h, --help             Show this message.
   
EXAMPLES:
   Prompt for host names:
     install-hadoop2.sh -i
     install-hadoop2.sh --interactive
   
   Use values from files in the same directory:
     install-hadoop2.sh -f
     install-hadoop2.sh --file
             
EOF
}

while true;
do
  case "$1" in

    -h|--help)
      help
      exit 0
      ;;
    -i|--interactive)
      interactive
      install
      shift
      ;;
    -f|--file)
      file
      install
      shift
      ;;
    --)
      shift
      break
      ;;
  esac
done

创新互联拥有网站维护技术和项目管理团队，建立的售前、实施和售后服务体系，为客户提供定制化的网站设计、成都网站制作、网站维护、乐山服务器托管解决方案。为客户网站安全和日常运维提供整体管家式外包优质服务。我们的网站维护服务覆盖集团企业、上市公司、外企网站、商城系统网站开发、政府网站等各类型客户群体，为全球数千家企业提供全方位网站维护、服务器维护解决方案。