011.分布式可视化DAG工作流任务调度系统DolphinScheduler-1.3.3安装部署

# 1. 解压源码包
[admin@bdc01 ~]$ unzip apache-dolphinscheduler-incubating-1.3.3-src.zip
[admin@bdc01 ~]$ cd apache-dolphinscheduler-incubating-1.3.3-src-release

# 2.修改顶层pom.xml，未列出的保持默认即可
<project xmlns="..." >
    <groupId>org.apache.dolphinscheduler</groupId>
    <artifactId>dolphinscheduler</artifactId>
    <!-- 定制化DolphinScheduler版本 -->
    <version>1.3.3-hdp3.1.4.0</version>
    <packaging>pom</packaging>
    
    <properties>
        <!-- 修改为hdp3.1的hadoop和hive的版本 -->
        <hadoop.version>3.1.1</hadoop.version>
        <hive.jdbc.version>3.1.0</hive.jdbc.version>
        ...
    </properties>
    
    <dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>mysql</groupId>
                <artifactId>mysql-connector-java</artifactId>
                <version>${mysql.connector.version}</version>
                # 把scope这一行删掉
                # <scope>test</scope>
            </dependency>
            ...
        </dependencies>
    </dependencyManagement>
    
    <build>
        <pluginManagement>
            <plugins>
                <plugin>
                    <groupId>org.apache.maven.plugins</groupId>
                    <artifactId>maven-javadoc-plugin</artifactId>
                    <version>${maven-javadoc-plugin.version}</version>
                    <configuration>
                        <source>8</source>
                        <failOnError>false</failOnError>
                        <!-- 添加这一行 -->
                        <additionalparam>-Xdoclint:none</additionalparam>
                    </configuration>
                </plugin>
                ...
            </plugins>
        </pluginManagement>
    </build>
    ...
</project> 

# 3.修改其他所有pom.xml文件，修改version
# 包括：
# dolphinscheduler-alert/pom.xml
# dolphinscheduler-api/pom.xml
# dolphinscheduler-common/pom.xml
# dolphinscheduler-dao/pom.xml
# dolphinscheduler-dist/pom.xml
# dolphinscheduler-plugin-api/pom.xml
# dolphinscheduler-remote/pom.xml
# dolphinscheduler-server/pom.xml
# dolphinscheduler-service/pom.xml
# dolphinscheduler-ui/pom.xml

<parent>
    <groupId>org.apache.dolphinscheduler</groupId>
    <artifactId>dolphinscheduler</artifactId>
    <version>1.3.3-hdp3.1.4.0</version>
</parent>

# 4.编译源码
[admin@bdc01 apache-dolphinscheduler-incubating-1.3.3-src-release]$ mvn -U -X clean package -Prelease -Dmaven.test.skip=true

# 5.获取安装包
# dolphinscheduler-dist/target目录下
apache-dolphinscheduler-incubating-1.3.3-hdp3.1.4.0-dolphinscheduler-bin.tar.gz

mysql> set global validate_password_policy=0;
mysql> set global validate_password_mixed_case_count=0;
mysql> set global validate_password_number_count=3;
mysql> set global validate_password_special_char_count=0;
mysql> set global validate_password_length=3;
mysql> CREATE DATABASE dolphinscheduler DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
mysql> CREATE USER 'ds'@'%' IDENTIFIED BY '123456';
mysql> GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'ds'@'%' IDENTIFIED BY '123456';
mysql> flush privileges;

######################### 1.修改conf/datasource.properties ######################### 
# 注释掉postgresql相关
# postgresql
#spring.datasource.driver-class-name=org.postgresql.Driver
#spring.datasource.url=jdbc:postgresql://localhost:5432/dolphinscheduler
#spring.datasource.username=test
#spring.datasource.password=test
# 修改为MySQL数据源
spring.datasource.driver-class-name=com.mysql.jdbc.Driver
# 需要修改ip，本机localhost即可spring.datasource.url=jdbc:mysql://hdp01:3306/dolphinscheduler?useUnicode=true&characterEncoding=UTF-8&allowMultiQueries=true
spring.datasource.username=ds
spring.datasource.password=123456

######################### 2.初始化数据库 ######################### 
[admin@hdp01 dolphinscheduler-1.3.3]$ sh script/create-dolphinscheduler.sh
......
22:48:56.305 [main] INFO org.apache.dolphinscheduler.dao.upgrade.shell.CreateDolphinScheduler - create DolphinScheduler success

######################### 3.修改conf/env/dolphinscheduler_env.sh #########################
export HADOOP_HOME=/usr/hdp/current/hadoop-client
export HADOOP_CONF_DIR=/etc/hadoop/conf
#export SPARK_HOME1=/opt/soft/spark1
export SPARK_HOME2=/usr/hdp/current/spark2-client
export PYTHON_HOME=/usr/bin/python
export JAVA_HOME=/usr/java/jdk
export HIVE_HOME=/usr/hdp/current/hive-client
# 这两个组价没有安装先注释掉
#export FLINK_HOME=/opt/soft/flink
#export DATAX_HOME=/opt/soft/datax/bin/datax.py
#export PATH=$HADOOP_HOME/bin:$SPARK_HOME1/bin:$SPARK_HOME2/bin:$PYTHON_HOME:$JAVA_HOME/bin:$HIVE_HOME/bin:$PATH:$FLINK_HOME/bin:$DATAX_HOME:$PATH
export PATH=$HADOOP_HOME/bin:$SPARK_HOME2/bin:$PYTHON_HOME:$JAVA_HOME/bin:$HIVE_HOME/bin:$PATH
# 添加执行权限
[admin@hdp01 dolphinscheduler-1.3.3]$ chmod u+x conf/env/dolphinscheduler_env.sh

######################### 4.修改conf/config/install_config.conf #########################
# 这里填 mysql or postgresql
dbtype="mysql"
# 数据库连接地址
dbhost="hdp01:3306"
# 数据库用户名
username="ds"
# 数据库名
dbname="dolphinscheduler"
# 数据库密码, 如果有特殊字符，请使用\转义
password="123456"
# Zookeeper地址
zkQuorum="hdp02:2181,hdp03:2181,hdp04:2181"
# 将DS安装到哪个目录，如: /opt/apps/dolphinscheduler，不同于现在的目录
installPath="/opt/apps/dolphinscheduler-1.3.3"
# 使用哪个用户部署
deployUser="admin"
# 邮件配置，以qq邮箱为例
# 邮件服务地址
mailServerHost="smtp.exmail.qq.com"
# 邮件服务端口
mailServerPort="25"
# mailSender和mailUser配置成一样即可
# 发送者
mailSender="xxxxxx"
# 发送用户
mailUser="xxxxxxxxxx"
# 邮箱密码
mailPassword="xxxxxxxxxx"
# TLS协议的邮箱设置为true，否则设置为false
starttlsEnable="true"
# 开启SSL协议的邮箱配置为true，否则为false。注意: starttlsEnable和sslEnable不能同时为true
sslEnable="false"
# 邮件服务地址值，参考上面 mailServerHost
sslTrust="smtp.exmail.qq.com"
# 业务用到的比如sql等资源文件上传到哪里，可以设置：HDFS,S3,NONE
# 单机如果想使用本地文件系统，请配置为HDFS，因为HDFS支持本地文件系统；
# 如果不需要资源上传功能请选择NONE。强调一点：使用本地文件系统不需要部署hadoop
resourceStorageType="HDFS"
# 如果上传资源保存想保存在hadoop上，hadoop集群的NameNode启用了HA的话
# 需要将hadoop的配置文件core-site.xml和hdfs-site.xml放到安装路径的conf目录下
# 本例即是放到/opt/apps/dolphinscheduler-1.3.3/conf下面，并配置namenode cluster名称
# 如果NameNode不是HA,则只需要将mycluster修改为具体的ip或者主机名即可
defaultFS="hdfs://hdp01:8020"
s3Endpoint="http://192.168.xx.xx:9010"
s3AccessKey="xxxxxxxxxx"
s3SecretKey="xxxxxxxxxx"
# 如果没有使用到Yarn,保持以下默认值即可;
# 如果ResourceManager是HA，则配置为ResourceManager节点的主备ip或者hostname,比如"192.168.xx.xx,192.168.xx.xx";
# 如果是单ResourceManager请配置yarnHaIps=""即可
yarnHaIps=""
# 如果ResourceManager是HA或者没有使用到Yarn保持默认值即可；
# 如果是单ResourceManager，请配置真实的ResourceManager主机名或者ip
singleYarnIp="hdp01"
# 资源上传根路径,主持HDFS和S3,由于hdfs支持本地文件系统，需要确保本地文件夹存在且有读写权限
resourceUploadPath="/dolphinscheduler"
# 具备权限创建resourceUploadPath的用户
hdfsRootUser="hdfs"
kerberosStartUp="false"
krb5ConfPath="$installPath/conf/krb5.conf"
keytabUserName="hdfs-mycluster@ESZ.COM"
keytabPath="$installPath/conf/hdfs.headless.keytab"
apiServerPort="12345"
# 在哪些机器上部署DS服务，本机选localhost
ips="hdp01,hdp02,hdp03,hdp04"
# ssh端口,默认22
sshPort="22"
# master服务部署在哪台机器上
masters="hdp01,hdp02"
# worker服务部署在哪台机器上,并指定此worker属于哪一个worker组,下面示例的default即为组名
workers="hdp01:default,hdp02:default,hdp03:default,hdp04:default"
# 报警服务部署在哪台机器上
alertServer="hdp03"
# 后端api服务部署在在哪台机器上
apiServers="hdp04"

######################### 5.修改conf/master.properties，未列出的保持默认即可 #########################
# 剩余内存不足此值（单位G）后，DS将不能执行任务，默认为物理内存的十分之一
master.reserved.memory=1
# CPU负载最大值，当达到此值后，DS将不能执行任务，默认为cpu核数*2，因为我的核数为2，所以我调到了这个值
master.max.cpuload.avg=10

######################### 6.修改conf/worker.properties，未列出的保持默认即可 #########################
worker.reserved.memory=1
worker.max.cpuload.avg=10

######################### 7.将core-site.xml和hdfs-site.xml文件拷贝到DS安装conf目录下 #########################
[admin@hdp01 dolphinscheduler-1.3.3]$ cp /etc/hadoop/conf/core-site.xml /etc/hadoop/conf/hdfs-site.xml conf/

######################### 8.修改bin/dolphinscheduler-daemon.sh和script/dolphinscheduler-daemon.sh #########################
# 两个文件保持一致
# 适当调整master-server的堆内存
elif [ "$command" = "master-server" ]; then
  HEAP_INITIAL_SIZE=1g # 默认4g
  HEAP_MAX_SIZE=2g # 默认4g
  HEAP_NEW_GENERATION__SIZE=1g # 默认2g
  LOG_FILE="-Dlogging.config=classpath:logback-master.xml -Ddruid.mysql.usePingMethod=false"
  CLASS=org.apache.dolphinscheduler.server.master.MasterServer
  
######################### 9.一键安装部署 #########################
[admin@hdp01 dolphinscheduler-1.3.3]$ sh install.sh

######################### 10.检查进程 #########################
# 各节点的进程与我规划的一致
[admin@hdp01 ~]$ jps
86181 WorkerServer
86228 LoggerServer
86125 MasterServer
[admin@hdp02 ~]$ jps
56663 MasterServer
56855 LoggerServer
56793 WorkerServer
[admin@hdp03 ~]$ jps
42624 WorkerServer
42739 AlertServer
42676 LoggerServer
[admin@hdp04 ~]$ jps
42932 WorkerServer
42987 LoggerServer
43067 ApiApplicationServer

bin/stop-all.sh

bin/start-all.sh

bin/dolphinscheduler-daemon.sh start master-server
bin/dolphinscheduler-daemon.sh stop master-server

bin/dolphinscheduler-daemon.sh start worker-server
bin/dolphinscheduler-daemon.sh stop worker-server

bin/dolphinscheduler-daemon.sh start api-server
bin/dolphinscheduler-daemon.sh stop api-server

bin/dolphinscheduler-daemon.sh start logger-server
bin/dolphinscheduler-daemon.sh stop logger-server

bin/dolphinscheduler-daemon.sh start alert-server
bin/dolphinscheduler-daemon.sh stop alert-server