CDH大数据平台搭建

一、准备4台虚拟机

ip	机器名	角色
192.168.160.128	host01	CM管理组件
192.168.160.130	host02	namenode/resource manager/hbase master/impala statestore/impala catalog server/hiveserver2/hive metastore/server/zookeeper server
192.168.160.129	host03	datanode/zookeeper server/node manager/hbase regionserver/impala daemon
192.168.160.131	host04	datanode/zookeeper server/node manager/hbase regionserver/impala daemon

 

二、相关包下载

CM：在集群中进行hadoop等大数据相关服务安装和监控管理的组件

下载地址： http://archive.cloudera.com/cm5/cm/5/

CDH：基于稳定版本的hadoop构建，可直接用于生产环境

下载地址： http://archive.cloudera.com/cdh5/parcels/5.7.0/

http://archive.cloudera.com/cdh5/parcels/5.7.0/manifest.json

CDH-5.7.0-1.cdh5.7.0.p0.45-el7.parcel、 CDH-5.7.0-1.cdh5.7.0.p0.45-el7.parcel.sha1、 manifest.json

三.准备工作

1.--vi /etc/sysconfig/network 修改主机名

hostnamectl  set-hostname host01

2.所有节点打通ssh设置ssh无密码登录

ssh-keygen -t rsa

cat ~/.ssh/id_rsa.pub  >> ~/.ssh/authorized_keys

scp id_rsa.pub root@192.168.160.128:/root/.ssh/02.pub

scp id_rsa.pub root@192.168.160.128:/root/.ssh/03.pub

scp id_rsa.pub root@192.168.160.128:/root/.ssh/04.pub

cat 02.pub >> authorized_keys

cat 03.pub >> authorized_keys

cat 04.pub >> authorized_keys

scp authorized_keys root@192.168.160.129:/root/.ssh/

scp authorized_keys root@192.168.160.130:/root/.ssh/

scp authorized_keys root@192.168.160.131:/root/.ssh/

3.所有节点安装oracle的java（运行cdh5需要oracle的jdk，java7的支持）

卸载自带openjdk：rpm -qa|grep java;rpm -e --nodeps 包名

下载jdk： http://archive.cloudera.com/cm5/redhat/7/x86_64/cm/5.7.0/RPMS/x86_64/

rpm -ivh oracle-j2sdk1.7-1.7.0+update67-1.x86_64.rpm

cd /usr/java

ln -s jdk1.7.0_67-cloudera latest

ln -s /usr/java/latest default

vi /etc/profile 在文件末尾配置java环境变量

export JAVA_HOME=/usr/java/jdk1.7.0_67-cloudera

export PATH=$PATH:$JAVA_HOME/bin

export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar

source /etc/profile

4.130上安装mysql

tar -zxvf mysql-5.6.48-linux-glibc2.12-x86_64.tar.gz

ln -s mysql-5.6.48-linux-glibc2.12-x86_64 mysql

cd /usr/local/mysql  mkdir logs

cp mysql/support-files/my-default.cnf /etc/my.cnf

编辑my.cnf

transaction_isolation = READ-COMMITTED

key_buffer_size = 32M

max_allowed_packet = 32M

thread_stack = 256K

thread_cache_size = 64

query_cache_limit = 8M

query_cache_size = 64M

query_cache_type = 1

max_connections = 550

log_bin = /usr/local/mysql/logs/mysql_binary_log

binlog_format = mixed

read_buffer_size = 2M

read_rnd_buffer_size = 16M

sort_buffer_size = 8M

join_buffer_size = 8M

innodb_file_per_table = 1

innodb_flush_log_at_trx_commit = 2

innodb_log_buffer_size = 64M

innodb_buffer_pool_size = 4G

innodb_thread_concurrency = 8

innodb_flush_method = O_DIRECT

innodb_log_file_size = 512M

explicit_defaults_for_timestamp

log_error = /usr/local/logs/mysqld.log

pid_file = /usr/local/mysql/data/mysqld.pid

sql_mode =   "STRICT_TRANS_TABLES,NO_ENGINE_SUBSTITUTION,NO_ZERO_DATE,NO_ZERO_IN_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER"

 

/usr/local/mysql

scripts/mysql_install_db --user=root

 

vi /etc/profile

export MYSQL_HOME=/usr/local/mysql

export PATH=$MYSQL_HOME/bin:$PATH

启动数据库：mysqld_safe --user=root &

修改root密码：mysqladmin -u root  password 

连接：mysql -u root -p

linux下mysql_secure_installation

wget http://dev.mysql.com/get/Downloads/Connector-J/mysql-connector-java-5.1.38.tar.gz

tar -zxvf mysql-connector-java-5.1.38.tar.gz  cp mysql-connector-java-5.1.38/mysql-connector-java-5.1.38-bin.jar /usr/share/java/mysql-connector-java.jar

创建数据库和用户

create database metastore DEFAULT CHARACTER SET latin1;

grant all on metastore.* TO 'hive'@'%' IDENTIFIED BY '123456';

flush privileges;

grant all on scm.* to 'root'@'%' identified by '123456' with grant option;

flush privileges;

5.关闭防火墙和SELinux

临时service iptables stop 重启生效chkconfig iptables off;setenforce 0,修改/etc/selinux/config下的SELINUX=disabled

6.配置系统内核参数

echo never > /sys/kernel/mm/transparent_hugepage/defrag  在/etc/rc.local文件中加入

echo "vm.swappiness = 0" >> /etc/sysctl.conf    sysctl -p生效

四、在192.168.160.128主节点解压安装CM

将CM包放在/opt下解压：tar xzf  cloudera-manager-centos7-cm5.7.0_x86_64.tar.gz

CM Agent配置

修改/opt/cm-5.7.0/etc/cloudera-scm-agent/config.ini中的主机名和端口号

五、同步主节点的cm-5.7.0文件夹到其他机器

scp -r /opt/cm-5.7.0  root@192.168.160.129:/opt/

scp -r /opt/cm-5.7.0  root@192.168.160.130:/opt/

scp -r /opt/cm-5.7.0  root@192.168.160.131:/opt/

六、所有节点创建cloudera-scm用户

useradd --system --home=/opt/cm-5.7.0/run/cloudera-scm-server --no-create-home --shell=/bin/false --comment "Cloudera SCM User" cloudera-scm

七、给ClouderaManager5建立数据库，存放CM的元数据

安装mysql的机器上scp mysql-connector-java-5.1.38-bin.jar root@192.168.160.128:/opt/cm-5.7.0/share/cmf/lib/

初始化CM5数据库：/opt/cm-5.7.0/share/cmf/schema/scm_prepare_database.sh mysql -h192.168.160.128 -uroot -p123456 --scm-host 192.168.160.128 scm 123456 123456

八、准备Parcels用来安装CHD5

在CMserver机器上mkdir -p /opt/cloudera/parcel-repo

chown cloudera-scm:cloudera-scm /opt/cloudera/parcel-repo

在每个节点：mkdir -p /opt/cloudera/parcels

chown cloudera-scm:cloudera-scm /opt/cloudera/parcels

CDH5相关包放到/opt/cloudera/parcel-repo，mv CDH-5.7.0-1.cdh5.7.0.p0.45-el7.parcel.sha1 CDH-5.7.0-1.cdh5.7.0.p0.45-el7.parcel.sha

 

CM相关启动脚本

启动服务端：/opt/cm-5.7.0/etc/init.d/cloudera-scm-server start

启动Agent服务：/opt/cm-5.7.0/etc/init.d/cloudera-scm-agent start

主节点启动CMserver和Agent，其他节点启动Agent

 

agent启动失败：INFO     SCM Agent Version: 5.7.0 Unable to create the pidfile

/opt/cm-5.7.0/run下创建目录 mkdir cloudera-scm-agent

九、安装

CDH5安装配置网址：http://192.168.160.128:7180/

admin/admin