数据采集+调度：cdh5.8.0+mysql5.7.17+hadoop+sqoop+hbase+oozie+hue

标签：用户   ppi   version   lis   引擎   profile   官方文档   set   auth   

 一、前期准备 

1.操作系统版本

# cat /etc/redhat-release
CentOS release 6.5 (Final)

# uname –a
Linux enc-bigdata05 2.6.32-431.el6.x86_64 #1 SMP Fri Nov 22 03:15:09 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

2.前期软件准备

2-1. jdk版本: jdk-7u80-linux-x64.tar.gz

[jdk1.7.80下载地址][1]

2-2. mysql版本: mysql-5.7.17-linux-glibc2.5-x86_64.tar.gz

[mysql-5.7.17-x86_64下载地址][2]

[mysql所有版本的下载地址][3]

2-3. cloudera manager版本: cloudera-manager-el6-cm5.8.0_x86_64.tar.gz

[cloudera-manager-el6-cm5.8.0_x86_64.tar.gz下载地址][4]

[cloudera-manager5所有版本的下载地址][5]

2-4. parecel包版本:

CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel
CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel.sha1
manifest.json

[CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel下载地址][6]

注意: 因为linux操作版本是centos6.5 x64,所以选择的cloudera manager和parecel包的版本都是对应el6版本，

       如果您的linux系统版本是centos7、ubuntu等等其他版本，更下载对应的cloudera manager和parecel版本进行安装；

[cloudera官网关于CM5.8.0的安装说明][7]

----------

 二、cm5.8.0安装前环境的配置 

1.修改集群主机名hostname

# vim /etc/sysconfig/network
:HOSTNAME=enc-bigdata05

重启服务器shutdown -r now，是配置文件network生效

剩下的主机依次修改hostname, enc-bigdata0[1-6]

----------

2. 修改系统ip和hostname的映射: /etc/hosts

2-1.修改映射关系

# vim /etc/hosts

2-2.将文件hosts复制发送到集群的每一个节点

# for a in {1..6} ; do scp /etc/hosts enc-bigdata0$a:/etc/hosts ; done

注意: 将localhost的本地映射关系删除，或者放在文件最后面

127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6

2-3. 检查是否集群每一个节点的hosts都已经修改过来了

# for a in {1..6} ; do ssh enc-bigdata0$a cat /etc/hosts ; done

----------
3.设置ssh免密码登入，打通集群的每一个节点

3-1.创建公钥和私钥

# ssh-keygen -t rsa -P ‘‘

注意: 在集群的每一台节点上运行上述命令，创建当前用户的公钥和私钥；

每一个用户都可以创建自己的ssh公钥和私钥，打通每一个节点，可以用其它用户(非root用户)来安装CDH5.8.0;

3-2.将每一个节点的公钥id_rsa.pub放入到认证文件中authorized_keys;

# for a in {1..6}; do ssh enc-bigdata0$a cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys; done

3-3.将认证文件authorized_keys复制发送到每一个节点上去: /root/.ssh/authorized_keys

# for a in {1..6}; do scp /root/.ssh/authorized_keys enc-bigdata0$a:/root/.ssh/authorized_keys ; done

----------

4.安装jdk1.7.0_80

4-1.解压jdk到指定路径下

# tar -zxvf jdk-7u80-linux-x64.tar.gz -C /usr/local/

4-2.将解压后的jdk1.7.0_80目录发送到每一台节点的指定目录下

# for a in {1..6}; scp -r /usr/local/jdk1.7.0_80/ enc-bigdata0$a:/usr/local/; done

4-3.修改配置文件 /etc/profile

# vim /etc/profile

#install jdk-7u80-linux-x64
export JAVA_HOME=/usr/local/jdk1.7.0_80
export JAVA_BIN=$JAVA_HOME/bin
export JAVA_LIB=$JAVA_HOME/lib
export CLASSPATH=.:$JAVA_LIB/tools.jar:$JAVA_LIB/dt.jar
export PATH=$JAVA_BIN:$PATH

4-4.将配置文件复制发送到每一个节点下

# for a in {1..6}; scp -r /etc/profile enc-bigdata0$a:/etc/profile; done

# source /etc/profile

# for a in {1..6}; do ssh enc-bigdata0$a /usr/local/jdk1.7.0_80/bin/java -version ; done

5.修改系统参数

5-1.关闭防火墙
# service iptables status/stop

# for a in {1..6}; do ssh enc-bigdata0$a /etc/init.d/iptables status; done

# for a in {1..6}; do ssh enc-bigdata0$a /etc/init.d/iptables stop; done

# for a in {1..6}; do ssh enc-bigdata0$a chkconfig iptables off; done

# for a in {1..6}; do ssh enc-bigdata0$a chkconfig iptables --list; done

5-2.关闭SELINUX

# setenforce 0 (临时生效)

# vim /etc/selinux/config (永久生效)

: SELINUX=disabled

# for a in {1..6}; scp /etc/selinux/config enc-bigdata0$a: /etc/selinux/config; done

5-3.修改swappiness

# echo "vm.swappiness=0" >> /etc/sysctl.conf (永久生效)

# echo 0 > /proc/sys/vm/swappiness (临时生效)

# for a in {1..6}; do ssh enc-bigdata0$a echo "vm.swappiness=0" >> /etc/sysctl.conf; done

5-4.编辑/etc/security/limits.conf

# vim /etc/security/limits.conf

* soft nofile 65535

* hard nofile 65535

# for a in {1..6}; scp /etc/security/limits.conf enc-bigdata0$a: /etc/security/limits.conf; done

5-5.修改/etc/rc.local (开机运行的命令)

#vim /etc/rc.local (永久生效)

: echo never > /sys/kernel/mm/transparent_hugepage/defrag

# for a in {1..6}; do ssh enc-bigdata0$a echo never > /sys/kernel/mm/transparent_hugepage/defrag ; done (临时生效)

三、安装mysql5.7.17 

1. 检查之前是否安装过mysql，删除之前安装的mysql

# rpm -qa | grep mysql

mysql-libs-5.1.71-1.el6.x86_64

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

# find / -name mysql

# rm -rf …

注意：卸载不了就强制卸载，--nodeps --force

2. 新增用户，指定家目录，指定登陆级别

# usedel mysql

# groupdel mysql

# mkdir /usr/local/mysql

# useradd -d /usr/local/mysql/ -s /sbin/nologin -m mysql

# chown -R mysql:mysql /usr/local/mysql

# chmod 755 /usr/local/mysql

3. 解压到制定路径下: /usr/local/mysql

# tar -zxvf mysql-5.7.17-linux-glibc2.5-x86_64.tar.gz -C /tmp/

# mv /tmp/mysql-5.7.17-linux-glibc2.5-x86_64/* /usr/local/mysql/

说明：将解压后的安装包里的文件拷贝到mysql中

# mkdir /usr/local/mysql/data

说明：创建数据存放路径

# chown -R mysql:mysql /usr/local/mysql/

4.安装和初始化

# /usr/local/mysql/bin/mysql_install_db --user=mysql --basedir=/usr/local/mysql/ --datadir=/usr/local/mysql/data/

# cp -a /usr/local/mysql/support-files/my-default.cnf /etc/my.cnf

# cp -a /usr/local/mysql/support-files/mysql.server /etc/init.d/mysqld

# /usr/local/mysql/bin/mysqld_safe --user=mysql &

5.启动mysqld

# /etc/init.d/mysqld status

# /etc/init.d/mysqld restart

# chkconfig mysqld on

# chkconfig mysqld --list

添加快捷方式：ln -s /usr/local/mysql/bin/mysql /usr/bin/

6.初始密码
mysql5.7会在root用户家目录下生成一个初始化密码。

# cat /root/.mysql_secret

Password set for user ‘root@localhost‘ at 2016-06-01 15:23:25 FhCmc,k50X/y

# mysql -uroot -p （输入初始化密码）

//修改密码
mysql> SET PASSWORD = PASSWORD(‘123qwe‘);

mysql> flush privileges;

7. 添加远程访问

第一种方法: 改表

mysql> use mysql;

mysql> update user set host = ‘%‘ where user = ‘root‘;

第二种方法: 授权

mysql>GRANT ALL PRIVILEGES ON *.* TO ‘root‘@‘%‘;

mysql>FLUSH PRIVILEGES

8. 问题: 忘记初始化密码

# service mysqld stop

# vim /etc/my.cnf 在末尾添加 skip-grant-tables

# service mysqld start

# mysql -uroot -p (可以免密码登入)

mysql> set password for ‘root‘@‘localhost‘=password(‘new password‘);

mysql>flush privileges;

9. 配置/etc/my.cnf

---------------------------------------------------------------------

[mysql]

# 设置mysql客户端默认字符集

default-character-set=utf8

[mysqld]

#设置3306端口

port = 3306

# 设置mysql的安装目录

basedir=/usr/local/mysql

# 设置mysql数据库的数据的存放目录

datadir=/usr/local/mysql/data

tmpdir=/usr/local/mysql/tmp

# 允许最大连接数

max_connections=2000

#服务端使用的字符集默认为8比特编码的latin1字符集

character-set-server=utf8

# 创建新表时将使用的默认存储引擎

default-storage-engine=INNODB

#设置不区分大小写(0：大小写敏感；1：大小写不敏感)

lower_case_table_names=1

#安装模式:update mysql.user set authentication_string=password(‘123qwe‘) where user=‘root‘;

#skip-grant-tables #忘记初始化密码

--------------------------------------------------------------------

10. 创建数据库

mysql>create database amon default charset utf8 collate utf8_general_ci; #activity monitor

mysql>create database hive default charset utf8 collate utf8_general_ci; #hive

mysql>create database oozie default charset utf8 collate utf8_general_ci; #oozie

mysql>create database hue default charset utf8 collate utf8_general_ci; #hue

11. 新建用户(也可以延用root用户)

mysql>create user ‘admin‘@‘%‘identified by ‘Admin@123‘;

mysql>grant all privileges on *.* to ‘admin‘@‘%‘;

mysql>flush privileges;

----------

四、安装cloudera manager 

1.解压cloudera-manager-el6-cm5.8.0_x86_64.tar.gz到指定路径:/opt

# tar -zxvf cloudera-manager-el6-cm5.8.0_x86_64.tar.gz -C /opt/

# for a in {1..6}; do scp -r /opt/cm-5.8.0/ enc-bigdata0$a:/opt/;done

2.修改配置文件

# vim /opt/cm-5.8.0/etc/cloudera-scm-agent/config.ini

server_host=enc-bigdata05

# for a in {1..6}; do scp -r /opt/cm-5.8.0/etc/cloudera-scm-agent/config.ini enc-bigdata0$a:/opt/cm-5.8.0/etc/cloudera-scm-agent/config.ini ; done

3.创建CDH内置数据库scm

# mv mysql-connector-java-5.1.40-bin.jar /opt/cm-5.7.0/share/cmf/lib

# /opt/cm-5.7.1/share/cmf/schemascm_prepare_database.sh mysql -hlocalhost -uroot –pAdmin@123 --scm-host localhost scm scm scm

scm_prepare_database.sh 数据库类型 服务器ip username password -scm-host [cloudera manager server所在服务器] scm scm scm;

4.将parecel包复制到cdh-server的指定目录下,修改CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel.sha1文件名

# cp CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel /opt/cloudera/parcel-repo/

# cp CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel.sha1 /opt/cloudera/parcel-repo/

# cp manifest.json/opt/cloudera/parcel-repo/

# mv /opt/cloudera/parcel-repo/CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel.sha1 /opt/cloudera/parcel-repo/CDH-5.8.0-1.cdh5.8.0.p0.42-el6.parcel.sha

4.启动cloudera-scm-server

# /opt/cm-5.8.0/etc/init.d/cloudera-scm-server start/status/stop

5.启动cloudera-scm-agent

# for a in {1..6}; do ssh enc-bigdata0$a /opt/cm-5.8.0/etc/init.d/cloudera-scm-agent start; done

6.问题:cloudera-scm-agent 启动失败: Unable to create the pidfile

原因: 无法创建 /opt/cm-5.8.0/run/cloudera-scm-agent

解决办法:

# mkdir /opt/cm-5.8.0/run/cloudera-scm-agent

# chown -R cloudera-scm:cloudera-scm /opt/cm-5.8.0/run/cloudera-scm-agent

7.进入网址: http://IP:7180/ (配置CDH5.8.0)

enc-bigdata0[1-6].enc.cn ##点击模式

注意: 修改jdk主目录很重要，以防止集群主机无法检测到jdk；

注意: 要保证所有检查都成功，以防止后面报错

注意 : hive的元数据库如果是安装在mysql中的话，需要将相应数据库驱动拷贝到hive家目录下的lib；

# mv mysql-connector-java-5.1.40-bin.jar /opt/cloudera/parcels/CDH-5.6.0-1.cdh5.6.0.p0.45/lib/hive/lib

将相应的数据库驱动拷贝到对应的服务家目录下，以防止数据库连接失败；

五、问题

1. 关于namenode的内存问题，namenode的内存如果不够的话，会导致namenode服务进程死掉

问题描述：can‘t open /opt/cm-5.7.0/run/cloudera-scm-agent/process/1033-hdfs-NAMENODE/supervisord.conf , Permission deny

解决办法：增加内存

2. 无法检测到jdk存在

解决办法:设置JDK主目录

3. oozie web 页面不能显示

解决办法：上传ext-2.2.zip到指定路径：/var/lib/oozie

六、总结 

**本篇章仅供参考，具体的请以官方文档为准，谢谢！！**

数据采集+调度：cdh5.8.0+mysql5.7.17+hadoop+sqoop+hbase+oozie+hue

标签：用户   ppi   version   lis   引擎   profile   官方文档   set   auth   

原文地址：http://www.cnblogs.com/gusuchen/p/6661633.html