Cloudera Manager大数据平台搭建

Posted

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Cloudera Manager大数据平台搭建相关的知识，希望对你有一定的参考价值。

    在企业里快速搭建大数据平台除了Ambria外还可以用Cloudera Manager，这两种工具相对于纯手工搭建确实便捷很多，但是有利也有弊，相对于新手来说就不便于掌握内部原理，不好排错等,而纯手工搭建，出错较多，不容易成功。

一. 规划：
192.168.3.201 server/agent mysql
192.168.3.202 agent namenode
192.168.3.203 agent namenode（备） resourcemanager
192.168.3.204 agent resourcemanager （备）
192.168.3.205 agent

二. 环境准备

  1. 系统：Centos 6.5 64位

  2. 配置IP，hostname,
    各个服务器IP配置：# vim /etc/sysconfig/network-script/ifcfg-eth0
    service network restart
    各个服务器名配置：# vim /etc/sysconfig/network

    vim /etc/hosts

                192.168.3.201    rac201.com
            192.168.3.202    rac202.com
            192.168.3.203    rac203.com
            192.168.3.204    rac204.com
            192.168.3.205    rac205.com

  3.检查DNS配置，确保能上网
    # vim /etc/resolve.conf  
    #   ping www.baidu.com

  4.关闭防火墙和禁用selinux
     getenforce   查看selinux的状态并修改为disable状态
    #  vim  /etc/sysconfig/selinux
                 SELINUX=disable
            重启生效

  5. 禁用和关闭防火墙
      临时禁用  #  iptables -F    
                #  setenforce 0
            或设置开机不启动
                #   service  iptables stop
                #   chkconfig iptables off

  6. 所有服务器卸载自带的JDK，安装新版本，比如1.7或1.8版本的
     # rpm  -qa | grep java
     # rpm -e --nodeps tzdata-java-2012j-1.el6.noarch java-1.6.0-openjdk-1.6.0.0-1.50.1.11.5.el6_3.x86_64 java-1.7.0-openjdk-1.7.0.9-2.3.4.1.el6_3.x86_64
     # mkdir -p /opt/modules    
     # tar zxf jdk-8u141-linux-x64.gz -C /opt/modules   

  7. 对所有服务器配置JAVA环境变量
     # vi /etc/profile
     export JAVA_HOME=/opt/modules/jdk1.8.0_141
     export PATH=$PATH:$JAVA_HOME/bin
     # source /etc/profile
  8. 拷贝配置到其他服务器
     # scp -r  /opt/modules/jdk1.8.0_141  [email protected]:/opt/modules/
     # scp -r  /opt/modules/jdk1.8.0_141  [email protected]:/opt/modules/
     # scp -r  /opt/modules/jdk1.8.0_141  [email protected]:/opt/modules/
     # scp -r  /opt/modules/jdk1.8.0_141  [email protected]:/opt/modules/

     检查：  java -version

  9. 配置服务器之间免密钥登录
     9.1  #  ssh-keygen
          #  ssh-copy-id -i ~/.ssh/id_rsa.pub rac201.com
          #  ssh-copy-id -i ~/.ssh/id_rsa.pub rac202.com
          #  ssh-copy-id -i ~/.ssh/id_rsa.pub rac203.com
          #  ssh-copy-id -i ~/.ssh/id_rsa.pub rac204.com
          #  ssh-copy-id -i ~/.ssh/id_rsa.pub rac205.com

     9.2  # ssh rac01.com
          # exit            #不要忘记连接后要退出
          # ssh rac02.com
          # exit
          # ssh rac03.com
          # exit
          # ssh rac04.com
          # exit
          # ssh rac05.com
          # exit

 10. 服务器时间同步
    10.1 启动ntpd进程服务
        # service ntpd start
        # chkconfig ntpd on
        # chkconfig --list | grep ntpd      
    10.2 修改时间服务器的配置
        # vim /etc/ntp.conf
        打开注释，修改成本服务器的网段
        # restrict 192.168.3.0 mask 255.255.255.0 nomodify notrap               
        注释掉以下几行，可能是三行也可能是五行
        #server 0.centos.pool.ntp.org
        #server 1.centos.pool.ntp.org
        #server 2.centos.pool.ntp.org

        手工添加以下两行
        server  127.127.1.0     # local clock
        fudge   127.127.1.0 s   
        启动并且配置开机时启动
        # service ntpd start && chkconfig ntpd on && chkconfig --list | grep ntpd

11. 对所有服务器设置能打开的文件最大数量、进程最大数，能使用的内存等
     # vi /etc/security/limits.conf             
        *               soft    nofile          32728
        *               hard    nofile          1024567
        *               soft    nproc           65535
        *               hard    nproc           unlimited
        *               soft    memlock         unlimited
        *               hard    memlock         unlimited

     复制limits.conf到其他服务器中
     # scp /etc/security/limits.conf [email protected]:/etc/security/limits.conf
     # scp /etc/security/limits.conf [email protected]:/etc/security/limits.conf
     # scp /etc/security/limits.conf [email protected]:/etc/security/limits.conf
     # scp /etc/security/limits.conf [email protected]:/etc/security/limits.conf
     重启生效：  ulimit -a

三. Mysql数据库安装

  1. 在rac201服务器检查并卸载系统自动安装的Mysql版本,

    rpm -qa | grep mysql

        # rpm -e --nodeps mysql-libs-5.1.71-1.el6.x86_64

  2. 采用yum源方式安装Mysql5.6, 先下载yum源文件,

    wget http://repo.mysql.com/mysql57-community-release-el6-8.noarch.rpm

        # rpm -Uvh mysql57-community-release-el6-8.noarch.rpm
    修改yum源文件：
      vim /etc/yum.repos.d/mysql-community.repo
      [mysql56-community]
      enabled=1
      gpgcheck=1

      [mysql57-community]
      enabled=0
      gpgcheck=1

      vim /etc/yum.repos.d/mysql-community-source.repo
      [mysql56-community-source]
      enabled=1   
      gpgcheck=1

      [mysql-tools-preview-source]
      name=MySQL Tools Preview - Source
      baseurl=http://repo.mysql.com/yum/mysql-tools-preview/el/6/SRPMS
      enabled=1   
      gpgcheck=1

    用yum repolist mysql这个命令查看一下是否已经有mysql可安装文件
      #yum repolist all | grep mysql

    yum安装mysql, 一直yes下去
    # yum -y install mysql-community-server

    安装成功后，启动并初始化
    # service mysqld start
    # service mysqld status

    mysql刚安装完，mysql的root用户的密码默认是空的，需要给root用户设置密码密码，执行下列命令进行安全设置
    # mysql_secure_installation        

    输入当前用户：回车

    输入root密码：123456

    production environment.

    Remove anonymous users? [Y/n] y
     ... Success!

    Normally, root should only be allowed to connect from ‘localhost‘.  This
    ensures that someone cannot guess at the root password from the network.

    Disallow root login remotely? [Y/n] n
     ... skipping.

    By default, MySQL comes with a database named ‘test‘ that anyone can
    access.  This is also intended only for testing, and should be removed
    before moving into a production environment.

    Remove test database and access to it? [Y/n] n
     ... skipping.

    Reloading the privilege tables will ensure that all changes made so far
    will take effect immediately.

    Reload privilege tables now? [Y/n] y
     ... Success!

四. Cloudera Manager安装

  1. 在每台服务器上安装第三方依赖包，或者在Centos系统安装时安装好开发包
    yum -y install chkconfig python bind-utils psmisc libxslt zlib sqlite cyrus-sasl-plain cyrus-sasl-gssapi fuse portmap fuse-libs redhat-lsb

  2. 提前下载好cloudera-manager-el6-cm5.3.6_x86_64.tar.gz
    http://archive.cloudera.com/cm5/cm/5/cloudera-manager-el6-cm5.3.6_x86_64.tar.gz

  3. 在rac201服务器上创建CM的解压目录

    mkdir -p /opt/cloudera-manager

    tar -zxf cloudera-manager-el6-cm5.3.6_x86_64.tar.gz -C /opt/cloudera-manager

    在rac201服务器上修改修改agent配置

    vi /opt/cloudera-manager/cm-5.3.6/etc/cloudera-scm-agent/config.ini

    Hostname of the CM server.

    server_host=rac201.com 
server_port=7182

  4. 拷贝解压后的CM文件夹到其他服务器

    scp -r /opt/cloudera-manager/ [email protected]:/opt/

    # scp -r /opt/cloudera-manager/ [email protected]:/opt/
# scp -r /opt/cloudera-manager/ [email protected]:/opt/
# scp -r /opt/cloudera-manager/ [email protected]:/opt/

  5. 在所有服务器上创建CM的用户

    useradd --system --home=/opt/cloudera-manager/cm-5.3.6/run/cloudera-scm-server --no-create-home --shell=/bin/false cloudera-scm

    # cat /etc/passwd | grep cloudera-scm

  6. 设置CM数据库的权限

    mysql -uroot -p123456

    mysql> grant all privileges on *.*  to ‘temp‘@‘%‘ identified by ‘temp‘ with grant option;
mysql> grant all privileges on *.* to ‘temp‘@‘rac01.com‘ identified by ‘temp‘ with grant option;
mysql> flush privileges;
mysql> use  mysql;
mysql> select host ,user from user;     

进入rac201服务器以下文件夹，执行脚本命令，事先要将mysql-connector-java.jar放入/usr/share/java文件夹中
# cd /opt/cloudera-manager/cm-5.3.6/share/cmf/schema
# ./scm_prepare_database.sh mysql -h rac01.com -utemp -ptemp --scm-host rac01.com scm scm scm

  7. 提前下载好CDH源文件
    CDH-5.3.6-1.cdh5.3.6.p0.11-el6.parcel
    CDH-5.3.6-1.cdh5.3.6.p0.11-el6.parcel.sha
    在rac201服务器上创建以下目录

    mkdir -p /opt/cloudera/parcel-repo

    # mv CDH-5.3.6-1.cdh5.3.6.p0.11-el6.parcel* /opt/cloudera/parcel-repo/

在所有服务器上创建
# mkdir -p /opt/cloudera/parcels        

更改parcel目录的所有人
在rac201服务器上 
# chown -R cloudera-scm:cloudera-scm /opt/cloudera/parcel-repo/
在所有服务器上 
# chown -R cloudera-scm:cloudera-scm /opt/cloudera/parcels

  8. 启动CM的进程
    在rac201服务器上

    cd /opt/cloudera-manager/cm-5.3.6/etc/init.d

     # ./cloudera-scm-server  start

    在所有服务器上启动agent进程

    cd /opt/cloudera-manager/cm-5.3.6/etc/init.d/

     # ./cloudera-scm-agent start

    CM的服务器登录界面
    http://rac01.com:7180
    登录账号和密码 admin admin

技术分享图片

技术分享图片

技术分享图片

以上是关于Cloudera Manager大数据平台搭建的主要内容，如果未能解决你的问题，请参考以下文章

CDH大数据平台 26Cloudera Manager Console之sqoopazkaban调度作业实战(markdown新版)

CDH大数据平台 19Cloudera Manager Console之azkaban安装编译配置(markdown新版)

CDH大数据平台 18Cloudera Manager Console之Sentry权限kafka测试(markdown新版)

CDH大数据平台 31Cloudera Manager Console之impala hive负载均衡(markdown新版)

CDH使用秘籍：Cloudera Manager和Managed Service的数据库

CDH大数据平台 30Cloudera Manager Console之superset之redisldap配置(markdown新版四)