在实践了CDH6的离线部署之后,发现Cloudera的官方教程是基于CDH5.10的。同时,CDH6的内存占用比较高,我的机器有点遭不住,所以又把CDH5的部署实施走了一遍。

模板部署

  1. 安装CentOS7.10,并完成yum升级;
  2. 所有主机时间与阿里云时间服务器同步;
  3. 关闭selinux和firewalld;
  4. 安装JDK和mysql-connect;
[root@localhost ~]# wget https://repo.huaweicloud.com/java/jdk/8u202-b08/jdk-8u202-linux-x64.rpm
[root@localhost ~]# yum localinstall jdk-8u202-linux-x64.rpm
[root@localhost ~]# mv mysql-connector-java-5.1.39-bin.jar /usr/share/java/mysql-connector-java.jar 
  1. 关闭透明大页;
[root@localhost ~]# vim /etc/rc.d/rc.local

# 在最后一行添加如下内容
if test -f /sys/kernel/mm/transparent_hugepage/enabled; then
echo never > /sys/kernel/mm/transparent_hugepage/enabled
fi

if test -f /sys/kernel/mm/transparent_hugepage/defrag; then
echo never > /sys/kernel/mm/transparent_hugepage/defrag
fi

# 赋予rc.local具有可执行权限
[root@localhost ~]# chmod +x /etc/rc.d/rc.local
[root@localhost ~]# sh /etc/rc.d/rc.local

# 检查
[root@localhost ~]# cat /sys/kernel/mm/transparent_hugepage/defrag
always madvise [never]
[root@localhost ~]# cat /sys/kernel/mm/transparent_hugepage/enabled
always madvise [never]
  1. 调整swap策略
[root@localhost ~]# echo 10 > /proc/sys/vm/swappiness
[root@localhost ~]# echo vm.swappiness = 10 >> /etc/sysctl.conf
  1. 修改文件句柄数
[root@localhost ~]# vim /etc/security/limits.conf
# 添加如下内容
* soft nofile 100000
* hard nofile 100000
  1. 收尾
[root@localhost ~]# sys-unconfig
  1. 下载CM和CDH

CM下载地址 CDH下载地址

Agent部署

  1. 配置主机ssh信任
  2. 安装ansible
# 安装
[root@elephant ~]# yum install -y ansible

[root@elephant ~]# ls
CDH-5.16.2-1.cdh5.16.2.p0.8-el5.parcel.sha1
CDH-5.16.2-1.cdh5.16.2.p0.8-el7.parcel
Mail
anaconda-ks.cfg
cloudera-manager-agent-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm
cloudera-manager-daemons-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm
cloudera-manager-server-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm
cloudera-manager-server-db-2-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm
enterprise-debuginfo-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm
hadoop.tar.gz
jdk-8u202-linux-x64.rpm
manifest.json
mysql-connector-java-5.1.39-bin.jar

[root@elephant ~]# mkdir client
[root@elephant ~]# mv cloudera-manager-daemons-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm client/
[root@elephant ~]# mv cloudera-manager-agent-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm client/
[root@elephant ~]# tar zcvf client.tar.gz client/
  1. 分发
[root@elephant ~]# ansible all -m copy -a 'src=/root/client.tar.gz dest=/root/'
[root@elephant ~]# ansible all -a 'tar zxf /root/client.tar.gz'
  1. 安装
[root@elephant ~]# ansible all -a 'yum localinstall /root/client/cloudera-manager-daemons-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm -y'
[root@elephant ~]# ansible all -a 'yum localinstall /root/client/cloudera-manager-agent-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm -y'

# 修改配置文件
[root@elephant ~]# sed -i ':s/localhost/lion/g' /etc/cloudera-scm-agent/config.ini
[root@elephant ~]# ansible lion,tiger,horse,monkey -m copy -a 'src=/etc/cloudera-scm-agent/config.ini dest=/etc/cloudera-scm-agent/'

# 重启服务
[root@elephant ~]# ansible all -a 'systemctl enable cloudera-scm-agent --now'
[root@elephant ~]# ansible all -a 'systemctl restart cloudera-scm-agent'

管理主机部署

  1. 安装
[root@elephant ~]# ansible lion -m copy -a 'src=/root/cloudera-manager-server-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm dest=/root/'
[root@elephant ~]# ansible lion -m copy -a 'src=/root/CDH-5.16.2-1.cdh5.16.2.p0.8-el7.parcel dest=/opt/cloudera/parcel-repo'
[root@elephant ~]# ansible lion -m copy -a 'src=/root/manifest.json dest=/opt/cloudera/parcel-repo'
[root@elephant ~]# ansible lion -a 'yum localinstall /root/cloudera-manager-server-5.16.2-1.cm5162.p0.7.el7.x86_64.rpm -y'
  1. 数据库部署
[root@elephant ~]# ansible lion -a 'yum install -y mariadb mariadb-server'
[root@elephant ~]# ssh lion
[root@lion ~]# vim /etc/my.cnf.d/server.cnf
[mysqld]
key_buffer = 16M
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
server_id=1

binlog_format = mixed
read_buffer_size = 2M
read_rnd_buffer_size = 16M
sort_buffer_size = 8M
join_buffer_size = 8M

# InnoDB settings
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

[root@lion ~]# systemctl enable mariadb --now

vim cdh.sql

CREATE DATABASE scm DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON scm.* TO 'scm'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE amon DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON amon.* TO 'amon'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE rman DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON rman.* TO 'rman'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE hue DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hue.* TO 'hue'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE metastore DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON metastore.* TO 'hive'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE sentry DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON sentry.* TO 'sentry'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE nav DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON nav.* TO 'nav'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE navms DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON navms.* TO 'navms'@'%' IDENTIFIED BY 'passwd';
CREATE DATABASE oozie DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON oozie.* TO 'oozie'@'%' IDENTIFIED BY 'passwd';

# 导入数据库
[root@lion ~]# mysql -uroot -p < cdh.sql

# 扩展数据库架构和CDH6有区别
[root@lion ~]# cp /usr/share/java/mysql-connector-java.jar /usr/share/cmf/lib/
[root@lion ~]# sh /usr/share/cmf/schema/scm_prepare_database.sh mysql scm scm
Enter SCM password:
JAVA_HOME=/usr/java/jdk1.8.0_202-amd64
Verifying that we can write to /etc/cloudera-scm-server
Creating SCM configuration file in /etc/cloudera-scm-server
Executing:  /usr/java/jdk1.8.0_202-amd64/bin/java -cp /usr/share/java/mysql-connector-java.jar:/usr/share/java/oracle-connector-java.jar:/usr/share/java/postgresql-connector-java.jar:/usr/share/cmf/schema/../lib/* com.cloudera.enterprise.dbutil.DbCommandExecutor /etc/cloudera-scm-server/db.properties com.cloudera.cmf.db.
[                          main] DbCommandExecutor              INFO  Successfully connected to database.
All done, your SCM database is configured correctly!

# 启动服务
[root@lion parcel-repo]# cd /opt/cloudera/parcel-repo/
[root@lion parcel-repo]# sha1sum CDH-5.16.2-1.cdh5.16.2.p0.8-el7.parcel  awk '{ print $1 }' > CDH-5.16.2-1.cdh5.16.2.p0.8-el7.parcel.sha
[root@lion ~]# chkconfig cloudera-scm-server on
[root@lion ~]# systemctl start cloudera-scm-server

# 检查结果
[root@lion ~]# netstat -tlnp grep 7180
tcp        0      0 0.0.0.0:7180            0.0.0.0:*               LISTEN      3907/java 

GUI安装

接下来的安装就相对简单了,基本比CDH6简单。 start start start start start start start start start start start end