大数据环境安装部署步骤

Posted Blue眼泪2016

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了大数据环境安装部署步骤相关的知识,希望对你有一定的参考价值。

系统环境搭建:

系统环境初始配置:
1.配置IP
vi /etc/sysconfig/network-scripts/ifcfg-eth0

DEVICE="eth0"
BOOTPROTO="none"
HWADDR="00:0C:29:54:E0:EF"
IPV6INIT="yes"
NM_CONTROLLED="yes"
ONBOOT="yes"
TYPE="Ethernet"
IPADDR=192.168.0.18
NETMASK=255.255.255.0
GATEWAY=192.168.0.1
UUID="3a66fd1d-d033-411b-908d-7386421a281b"

2.配置hosts
    vi /etc/hosts

2.命名节点名称
    vi /etc/sysconfig/network

4.禁用防火墙
    service iptables stop 
    chkconfig iptables off

5.配置root用户ssh免密登录
    su - root
    cd ~
    #对每个节点分别产生公钥和私钥:
    cd  ~/.ssh
    ssh-keygen -t dsa -P ‘‘ -f ~/.ssh/id_dsa
    #将公钥文件复制保存为authorized_keys
    cat id_dsa.pub >> authorized_keys
    
6.修改vim
    cd ~
    vi .bashrc
#添加如下内容:
    alias vi=vim
#执行命令
    source .bashrc

环境变量配置【预先配置好】

6.先把所有环境变量配好
    cd /usr/local/
    rm -rf *
    vi /etc/profile
#添加如下内容最后执行source /etc/profile命令:
#配置java环境变量
export JAVA_HOME=/usr/local/jdk
export JAVA_BIN=$JAVA_HOME/bin
export JAVA_LIB=$JAVA_HOME/lib
export CLASSPATH=.:$JAVA_LIB/tools.jar:$JAVA_LIB/dt.jar

#配置Python环境变量
export PYTHON_HOME=/usr/local/python2.7


#配置HADOOP环境变量
export HADOOP_HOME=/usr/local/hadoop
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib"
export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib"

#配置HIVE环境变量
export HIVE_HOME=/usr/local/hive

#配置zookeeper环境变量
export ZOOKEEPER_HOME=/usr/local/zookeeper

#配置hbase的环境变量
export HBASE_HOME=/usr/local/hbase

#export PATH=.:$HBASE_HOME/bin:$ZOOKEEPER_HOME/bin:$HIVE_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_BIN:$PATH
export PATH=.:$HBASE_HOME/bin:$ZOOKEEPER_HOME/bin:$HIVE_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_BIN:$PYTHON_HOME/bin:$PATH


#设置时区
TZ=Asia/Shanghai
export TZ

    source /etc/profile
    

Python安装升级:

cd /software
    cp Python-2.7.13.tgz /usr/local/
    cp get-pip.py /usr/local/
    cd /usr/local/
    tar -zxvf Python-2.7.13.tgz
    cd /usr/local/Python-2.7.13
    mkdir /usr/local/python2.7
    ./configure --prefix=/usr/local/python2.7
    make && make install
    mv /usr/bin/python /usr/bin/python_old
    ln -s /usr/local/python2.7/bin/python2.7 /usr/bin/python
    #验证
    python -V
    cd /usr/bin/
    vi yum 
    #修改第一行为:
    #!/usr/bin/python_old
    
    cd /software
    cp pip-9.0.1.tar.gz /usr/local
    cd /usr/local 
    tar -zxvf pip-9.0.1.tar.gz
    yum install openssl-devel
    yum install ncurses-devel
    yum install sqlite-devel
    yum install zlib-devel
    yum install bzip2-devel
    yum install python-devel
    yum -y install zlib* 
    yum install zlibc zlib1g-dev
    yum install zlib
    yum install zlib-devel
    cd /usr/local
    wget http://pypi.python.org/packages/source/s/setuptools/setuptools-2.0.tar.gz --no-check-certificate
    tar zxvf setuptools-2.0.tar.gz
    cd setuptools-2.0
    python setup.py build
    python setup.py install
    python get-pip.py
    
    pip install package
    numpy:数据处理;
    
    matplotlib:数据可视化;
    
    pandas:数据分析;
    
    pyserial:串口通信(注:利用pip安装的pyserial是版本3的,而在WinXP上只能使用pyserial2版本);
    
    cx_freeze:将py打包成exe文件;
    
    pip install pyquery
    pip install beautifulsoup
    pip install numpy
    pip install matplotlib
    pip install pandas:数据分析;
    pip install pandas
    pip install cx_freeze
    pip install lrd
    pip install pyserial
    pip install nltk
    pip install mlpy
    pip install Pygame
    pip install Sh
    pip install Peewee
    pip install Gooey
    pip install pillow
    pip install xlrd
    pip install lxml
    pip install configparser
    pip install uuid
    pip install msgpack-python
    pip install psutil
    pip installmysql-Python
    pip install MySQL-Python
    pip install pymongo
    pip install cxOracle
    pip install Arrow
    pip install when.py
    pip install PIL
    pip install Pyquery
    pip install virtualenv

jdk安装

#-------------------------jdk安装-----------------------------
1.将jdk安装包jdk-8u51-linux-x64.gz上传至/usr/local/目录下
    cd /software
    cp jdk-8u51-linux-x64.gz /usr/local
    cd /usr/local
2.解压jdk安装包
    tar  -zxvf  jdk-8u51-linux-x64.gz
    #重命名安装包
    mv jdk1.8.0_51 jdk
3.配置环境变量[前面已经配置好]
    vi /etc/profile
    export JAVA_HOME=/usr/local/jdk
    export JAVA_BIN=$JAVA_HOME/bin
    export JAVA_LIB=$JAVA_HOME/lib
    export CLASSPATH=.:$JAVA_LIB/tools.jar:$JAVA_LIB/dt.jar
    #最后将所有的路径加到
    export PATH=.:$JAVA_BIN:$PATH
    #使环境变量生效
    source /etc/profile
4.验证jdk安装是否成功
    java -version
    

mysql安装

#-----------------------安装mysql------------------------
1.上传mysql安装包到/usr/local目录下
    cd /software
    cp MySQL-server-5.6.22-1.el6.x86_64.rpm MySQL-client-5.6.22-1.el6.x86_64.rpm /usr/local/
    cd /usr/local
2.卸载依赖包
    #查找安装了的mysql
    rpm -qa | grep mysql
    #如果有,则执行命令卸载
    rpm -e  mysql-libs-5.1.71-1.el6.x86_64  --nodeps
2.安装mysql
    rpm -ivh MySQL-client-5.6.22-1.el6.x86_64.rpm  --nodeps
    rpm -ivh MySQL-server-5.6.22-1.el6.x86_64.rpm  --nodeps
3.启动mysql服务
    service  mysql start
4.查看root账号密码并登陆
    cat /root/.mysql_secret   #CQslM7ZtrjTbwiFv
    #登录mysql
    mysql -uroot -p密码
    #设置密码
    mysql> SET PASSWORD = PASSWORD(root);
    #测试新密码登录
    mysql -uroot -proot 
5设置允许远程登录
    GRANT ALL PRIVILEGES ON *.* TO root@% IDENTIFIED BY root WITH GRANT OPTION;
    GRANT ALL PRIVILEGES ON *.* TO root@hadoop IDENTIFIED BY root WITH GRANT OPTION;
    flush privileges;
    exit;
6.设置开机自动启动
    chkconfig mysql on

安装hadoop

#------------------------hadoop安装--------------------------
1.上传hadoop安装包到/usr/local目录下
    cd /software
    cp hadoop-2.6.0.tar.gz hadoop-native-64-2.6.0.tar /usr/local
    cd /usr/local
2.解压hadoop安装包
    tar -xzvf  hadoop-2.6.0.tar.gz
    #重命令hadoop
    mv hadoop-2.6.0 hadoop 
3.设置hadoop环境变量
    vi /etc/profile
    export HADOOP_HOME=/usr/local/hadoop
    #修改:
    export PATH=.:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_BIN:$PATH 
4.配置hadoop的参数
#4.1 修改hadoop-env.sh文件
    #添加java_home的环境变量
    cd /usr/local/hadoop/etc/hadoop
    vi hadoop-env.sh
    JAVA_HOME=/usr/local/jdk
#4.2 配置core-site.xml
    cd /usr/local/hadoop/etc/hadoop
    vi core-site.xml
<configuration>
    <property>
        <name>fs.default.name</name>
        <value>hdfs://hadoop:9000</value>
        <description>change your own hostname</description>
    </property>
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/usr/local/hadoop/tmp</value>
    </property>  
</configuration>
  
#4.3 配置hdfs-site.xml  
  <configuration>
    <property>
        <name>dfs.replication</name>
        <value>1</value>
    </property>
    <property>
        <name>dfs.permissions</name>
        <value>false</value>
    </property>
  </configuration>
  
#4.4.配置mapred-site.xml
  <configuration>
    <property>
        <name>mapred.job.tracker</name>
        <value>hadoop:9001</value>
        <description>change your own hostname</description>
    </property>
</configuration>

9.64位系统错误问题处理
##安装Hadoop启动之后总有警告:
##Unable to load native-hadoop library for your platform... using builtin-Javaclasses where applicable
##这是因为在64位的linux系统中运行hadoop不兼容。
##这时候将准备好的64位的lib包解压到已经安装好的hadoop的lib目录下
#注意:是lib目录,而不是lib下的native目录
##执行如下命令:
#tar -x hadoop-native-64-2.4.0.tar -C  hadoop/lib/
    cd /usr/local
    cp hadoop-native-64-2.6.0.tar hadoop/lib
    cd /usr/local/hadoop/lib
    tar -xvf  hadoop-native-64-2.6.0.tar
#然后在环境变量中添加如下内容:
    vi /etc/profile
    export HADOOP_COMMON_LIB_NATIVE_DIR=$HADOOP_HOME/lib/native
    export HADOOP_OPTS="-Djava.library.path=$HADOOP_HOME/lib"
#最后使环境变量生效
    source  /etc/profile

5.对hadoop进行格式化
    hadoop namenode -format
6.启动hadoop
    start-all.sh
7.验证hadoop是否安装成功:
 输入命令:jps 
 #发现有五个java进程:
    DataNode
    NameNode
    SecondaryNameNode
    JobTracker
    TaskTracker
 #通过浏览器查看:
    HDFS:
        hadoop:50070 
    MapReduce:
        hadoop:50030 
 
8.修改windows下的文件,即可在本地电脑查看:
    C:\Windows\System32\drivers\etc\hosts

10.错误处理办法
    如果在windows中页面不能成功,有肯能
    NameNode进程启动没有成功?
    1.没有格式化
    2.配置文件
    3.hostname没有与ip绑定
    4.SSH的免密码登录没有配置成功
    #多次格式化也是错误的
    方法:删除/usr/local/hadoop/tmp文件夹,重新格式化

hive安装

1.上传hive安装包到/usr/local目录下
    cd /usr/local
2.解压hive安装包
    tar -zxvf hive-0.9.0.tar.gz
    mv hive-0.9.0 hive
3.配置hive环境变量
    vi /etc/profile
    export HIVE_HOME=/usr/local/hive
    export PATH=.:$HIVE_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_BIN:$PATH 
    source /etc/profile
4.上传驱动到/usr/local目录下并添加驱动到hive的lib目录下
    cd /usr/local
    cp mysql-connector-java-5.1.39-bin.jar  /usr/local/hive/lib/
5.hive安装参数配置
#修改hive-env.sh文件,添加hadoop的环境变量
    cd /usr/local/hive/conf
    cp hive-env.sh.template  hive-env.sh
    vi hive-env.sh
    export HADOOP_HOME=/usr/local/hadoop
#修改hive-log4j.properties文件
    cd /usr/local/hive/conf
    cp hive-log4j.properties.template hive-log4j.properties
    vi hive-log4j.properties
#log4j.appender.EventCounter=org.apache.hadoop.metrics.jvm.EventCounter
    log4j.appender.EventCounter=org.apache.hadoop.log.metrics.EventCounter

#修改hive-site.xml文件
    cd /usr/local/hive/conf
    cp hive-default.xml.template hive-site.xml
    vi hive-site.xml
#添加如下内容:
<property>
  <name>javax.jdo.option.ConnectionURL</name>
  <value>jdbc:mysql://hadoop:3306/hive?createDatabaseIfNotExist=true</value>
  <description>JDBC connect string for a JDBC metastore</description>
</property>
<property>
  <name>javax.jdo.option.ConnectionDriverName</name>
  <value>com.mysql.jdbc.Driver</value>
  <description>Driver class name for a JDBC metastore</description>
</property>
<property>
  <name>javax.jdo.option.ConnectionUserName</name>
  <value>root</value>
  <description>username to use against metastore database</description>
</property>
<property>
  <name>javax.jdo.option.ConnectionPassword</name>
  <value>root</value>
  <description>password to use against metastore database</description>
</property>
6.验证hive安装是否成功
 hive
 show databases;
 mysql -uroot -proot 
 show databases;
 use hive;
 show tables;
7.上传hive-builtins-0.9.0.jar包到hdfs的/usr/local/hive/lib/目录下
 cd /usr/local/hive/lib
 hdfs dfs -mkdir -p  /usr/local/hive/lib
 hdfs dfs -put hive-builtins-0.9.0.jar /usr/local/hive/lib

DB2安装

#--------------------------DB2安装-----------------------------------#
1.配置实例用户
    mkdir -p /db2home
    groupadd -g 600 edwgadm
    groupadd -g 601 edwgfenc
    groupadd -g 602 edwgdas
    useradd -u 600 -g 600 -d /home/edwinst edwinst
    useradd -u 601 -g 601 -d /db2home/edwfenc edwfenc
    useradd -u 602 -g 602 -d /db2home/edwdas edwdas
    passwd  edwinst
    passwd  edwfenc
    passwd  edwdas
2.创建目录
    mkdir -p /db2home
    mkdir -p /edwpath
    mkdir -p /edwpath/edwinst/NODE0000
    mkdir -p /edwp0
    mkdir -p /edwpath/edwinst/NODE0001
    mkdir -p /edwp1
    mkdir -p /edwpath/edwinst/NODE0002
    mkdir -p /edwp2
    mkdir -p /edwpath/edwinst/NODE0003
    mkdir -p /edwp3
    
    chown -R edwinst:edwgadm /edwpath
    chown -R edwinst:edwgadm /edwpath/edwinst/NODE0000
    chown -R edwinst:edwgadm /edwp0
    chown -R edwinst:edwgadm /edwpath/edwinst/NODE0001
    chown -R edwinst:edwgadm /edwp1
    chown -R edwinst:edwgadm /edwpath/edwinst/NODE0002
    chown -R edwinst:edwgadm /edwp2
    chown -R edwinst:edwgadm /edwpath/edwinst/NODE0003
    chown -R edwinst:edwgadm /edwp3
3.开启相关服务
    #yum install rpcbind nfs-utils
    #yum install xinetd
    service nfs restart
    service sshd restart
    #service portmap restart
    service rpcbind restart
    service xinetd restart
    chkconfig --level 2345 nfs on
    chkconfig --level 2345 nfslock on
    chkconfig --level 2345 sshd on
    #chkconfig --level 2345 portmap on
    chkconfig --level 2345 rpcbind on
    chkconfig --level 2345 xinetd on
创建实例:
    cd /opt/ibm/db2/V9.7/instance
     ./db2icrt -s ese -u edwfenc edwinst
1.关闭内存地址随机化机制
    vi /etc/sysctl.conf
    增加 kernel.randomize_va_space=0
    sysctl -p
2.上传db2安装包并解压安装
    cd /software
    cp v9.7fp9_linuxx64_server.tar.gz /usr/local
    cd /usr/local
    tar -xvf v9.7fp9_linuxx64_server.tar.gz
    cd ./server
    ./db2_install #开始安装db2,选择ESE企业版安装
    #安装ksh
    cd /software 
    rpm -ivh ksh-20120801-33.el6.x86_64.rpm
    su - edwinst 
    cd ~
    #对每个节点分别产生公钥和私钥:
    cd  ~/.ssh
    ssh-keygen -t dsa -P ‘‘ -f ~/.ssh/id_dsa
    #将公钥文件复制保存为authorized_keys
    cat id_dsa.pub >> authorized_keys
4.修改配置文件,添加端口号
    vi /etc/services
    db2c_edwinst    60000/tcp
    DB2_edwinst     60001/tcp
    DB2_edwinst_1   60002/tcp
    DB2_edwinst_2   60003/tcp
    DB2_edwinst_3   60004/tcp
    DB2_edwinst_4   60005/tcp
    DB2_edwinst_END 60006/tcp
5.新建实例
    cd /opt/ibm/db2/V9.7/instance
    ./db2icrt -s ese -u edwfenc edwinst
    su - edwinst
    vi /home/edwinst/sqllib/db2nodes.cfg(更新为一下内容)
    0 hadoop 0
    1 hadoop 1
    2 hadoop 2
    3 hadoop 3
6.设置管理参数:
    db2set DB2RSHCMD=/usr/bin/ssh
    db2set DB2CODEPAGE=1386
    db2set DB2_EXTENDED_OPTIMIZATION=ON
    db2set DB2_ANTIJOIN=YES
    db2set DB2COMM=TCPIP
    db2set DB2_PARALLEL_IO=*
    
    db2 update dbm cfg using SVCENAME db2c_edwinst
    db2start
7.创建EDW数据库
    db2 "CREATE DATABASE EDW AUTOMATIC STORAGE NO ON /edwpath USING CODESET GBK TERRITORY CN RESTRICTIVE"

    db2 connect to edw

    db2 "CREATE DATABASE PARTITION GROUP PDPG ON DBPARTITIONNUMS (0 to 3)"

    db2 "CREATE DATABASE PARTITION GROUP SDPG ON DBPARTITIONNUMS (0)"

    db2 "ALTER BUFFERPOOL IBMDEFAULTBP SIZE 20"

    db2 "CREATE BUFFERPOOL BP32K ALL DBPARTITIONNUMS SIZE 20 PAGESIZE 32K"

    db2 update db cfg using LOGFILSIZ  131072 LOGPRIMARY 30 LOGSECOND 5
    db2 update dbm cfg using FEDERATED YES

    db2 force application all
    db2stop
    db2start
    

9.检查并建立 BUFFERPOOL
  #db2 ALTER BUFFERPOOL IBMDEFAULTBP SIZE 250;
  db2 ALTER BUFFERPOOL IBMDEFAULTBP SIZE 20;
  #db2 CREATE BUFFERPOOL BP32K SIZE 16384 PAGESIZE 32768;
  #db2 CREATE BUFFERPOOL BP32K SIZE 50 PAGESIZE 32768;
  #db2 CONNECT RESET;
  db2 -x "select BPNAME,NPAGES,PAGESIZE from syscat.bufferpools with ur"
  
10.设置当前会话模式
    db2 set schema dainst
11.db2执行sql脚本
 在命令行中执行建表语句
 db2 -svtf crt_dwmm_etl_table.ddl

调度平台安装【另外一台机器】

#--------------------------调度平台安装【另外一台机器】-----------------------------------#

# 安装ncompress包
   yum -y install ncompress
# 上传版本程序包 etldata.tar.gz
# apache-tomcat-7.0.73.tar.gz job-schd-engine-0.1Silver.tar.gz JobSchd.war 至当前目录
    tar xf apache-tomcat-7.0.73.tar.gz -C ./etldata/script

    tar xf job-schd-engine-0.1Silver.tar.gz -C ./etldata/script

    cp JobSchd.war ./etldata/script/apache-tomcat-7.0.73/webapps
   
    cd /etl/etldata/script/apache-tomcat-7.0.73/webapps
    unzip JobSchd.war -d JobSchd
    cd /etl/etldata/script/apache-tomcat-7.0.73/webapps/JobSchd/WEB-INF/classes
    vi jdbc.properties
    
   /etl/etldata/script/apache-tomcat-7.0.73/webapps
 
    cd /etl/etldata/script/apache-tomcat-7.0.73/bin
    ./startup.sh
    浏览器访问:http://192.168.0.18:8080/JobSchd/logon.jsp
    用户:admin
    密码:12345678
    
创建联邦:
    db2 catalog tcpip node EDW remote hadoop server 60000
    db2 catalog db EDW as EDW at node EDW
    db2 terminate
    db2 connect to EDW user edwinst using edwinst
    db2 create wrapper drda
    db2 "create server EDW1SVR type DB2/UDB version 9.7 wrapper "drda" authorization \"edwinst\" password \"edwinst\" options(NODE ‘EDW‘,DBNAME ‘EDW‘)"
    db2 "create user mapping for \"dainst\" server EDW1SVR options(remote_authid ‘edwinst‘,remote_password ‘edwinst‘)"

 导入新增数据:
 import from /home/isinst/tmp/1.csvof del insert into XMETA.SRC_STU_INFO;
 
 导入替换数据s
 import from /home/isinst/tmp/1.csvof del replace into XMETA.SRC_STU_INFO;
 
DB2建表语句【例子】:
create table XMETA.SRC_STU_INFO (
    SRC_SYS_ID         INTEGER  NOT NULL    ,
    NAME             VARCHAR(20)  ,
    SCORE             INTEGER     ,
    SEX           VARCHAR(10)    NOT NULL   ); 
    
ALTER TABLE EXT_${tab_name} SET LOCATION hdfs://nameservice1/etldata/input/${tab_name}/${today}

 

以上是关于大数据环境安装部署步骤的主要内容,如果未能解决你的问题,请参考以下文章

大数据练习环境部署 -VMware及虚拟机安装

大数据系列之Hadoop分布式集群部署

大数据练习环境部署 - Hadoop集群部署

2021年大数据常用语言Scala:Scala开发环境安装

合成大西瓜开发源码,手把手教你运行和部署大西瓜游戏项目

streamsets安装部署