安装步骤

1、前置要求

2、软件包下载安装

wget https://mirrors.bfsu.edu.cn/apache/dolphinscheduler/1.3.4/apache-dolphinscheduler-incubating-1.3.4-dolphinscheduler-bin.tar.gz 下载dolphinscheduler-1.3.4版本。

注意：需要在集群中所有节点传输安装包并解压，完成集群分布式协调管理。

# 解压

tar -zxvf apache-dolphinscheduler-incubating-1.3.4-dolphinscheduler-bin.tar.gz -C /opt/dolphinscheduler;

# 改名

mv apache-dolphinscheduler-incubating-1.3.4-dolphinscheduler-bin  dolphinscheduler-bin

3、创建部署用户并确保hosts映射完备，以及其他配置

# 创建用户需使用root登录，设置部署用户名

useradd dolphinscheduler;

# 设置用户密码

echo "dolphinscheduler123" | passwd --stdin dolphinscheduler

# 配置sudo免密

echo 'dolphinscheduler  ALL=(ALL)  NOPASSWD: NOPASSWD: ALL' >> /etc/sudoers

sed -i 's/Defaults    requirett/#Defaults    requirett/g' /etc/sudoers

# 安装sshpass，需先安装epel

yum install -y epel-release

yum repolist

# 安装完成epel之后，就可以按照sshpass了

yum install -y sshpass

# 修改目录权限

dolphinscheduler-bin

4、数据库DB初始化

注意：因为使用的是CDH集群MySQL保存元数据，所以需要提供mysql-connector-java驱动包到dolphinscheduler的lib目录下：

mv mysql-connector-java-5.1.39.jar /opt/dolphinscheduler/dolphinscheduler-bin/lib

然后需要对数据库用户进行添加和权限修改：

# 添加dolphinscheduler用户

create user 'dolphinscheduler'@'%' identified by '123456';

# 创建dolphinscheduler数据库

CREATE DATABASE dolphinscheduler DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;

# 为dolphinescheduler赋予全IP权限

GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'dolphinscheduler'@'%' IDENTIFIED BY '123456';

# 刷新权限

flush privileges;

最后要对dolphinscheduler导入基础数据，首先需要对conf/datasource.properties文件进行以下内容的修改：

# postgresql

#spring.datasource.driver-class-name=com.mysql.jdbc.Driver

#spring.datasource.url=jdbc:mysql://10.0.2.138:3306/dolphinscheduler?characterEncoding=UTF-8&allowMultiQueries=true

spring.datasource.driver-class-name=com.mysql.jdbc.Driver

spring.datasource.url=jdbc:mysql://10.0.2.138:3306/dolphinscheduler?characterEncoding=UTF-8&allowMultiQueries=true

spring.datasource.username=dolphinscheduler

spring.datasource.password=123456

修改并保存后退出，然后执行script目录下的数据导入脚本：

sh script/create-dolphinscheduler.sh

至此，完成基础安装和数据导入。

5、修改运行参数

1】选装技术组件

当选装了技术组件如Hadoop、Spark、Flink、Hive等的时候，此时需要对conf/env/dolphinschedule_env.sh进行配置：

export HADOOP_HOME=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hadoop

export HADOOP_CONF_DIR=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hadoop/etc/hadoop

#export SPARK_HOME1=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/spark1

#export SPARK_HOME2=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/spark2

export SPARK_HOME=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/spark

#export PYTHON_HOME=/

export JAVA_HOME=/usr/java/jdk1.8.0_181

export HIVE_HOME=/opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hive

export FLINK_HOME=/opt/cloudera/parcels/FLINK-1.12.4-BIN-SCALA_2.11/lib/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_HOME/bin::$JAVA_HOME/bin:$HIVE_HOME/bin:$PATH:$FLINK_HOME/bin::$PATH

另外，安装了Hadoop和Hive，还需要将

HADOOPHOME/etc/hadoop下的core−site.xml和hdfs−site.xml，

HIVE_HOME/conf/hive-site.xml移动到目录/opt/dolphinscheduler下的新建conf目录中：

cp /opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hadoop/etc/hadoop/core-site.xml conf/

cp /opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hadoop/etc/hadoop/hdfs-site.xml conf/

cp /opt/cloudera/parcels/CDH-6.2.0-1.cdh6.2.0.p0.967373/lib/hive/conf/hive-site.xml conf

2】一键部署文件的修改

在conf/config/install_config.conf中有很多参数需要进行修改：

# NOTICE :  If the following config has special characters in the variable `.*[]^${}\+?|()@#&`, Please escape, for example, `[` escape to `\[`

# postgresql or mysql

dbtype="mysql"

# db config

# db address and port

dbhost="10.0.2.138:3306"

# db username

username="dolphinscheduler"

# database name

dbname="dolphinscheduler"

# db passwprd

# NOTICE: if there are special characters, please use the \ to escape, for example, `[` escape to `\[`

password="123456"

# zk cluster

zkQuorum="data-dev01:2181,data-dev02:2181,data-dev02:2181"

# Note: the target installation path for dolphinscheduler, please not config as the same as the current path (pwd)

installPath="/opt/dolphinscheduler"

# deployment user

# Note: the deployment user needs to have sudo privileges and permissions to operate hdfs. If hdfs is enabled, the root directory needs to be created by itself

deployUser="dolphinscheduler"

# alert config

# mail server host

mailServerHost="smtp.163.com"

# mail server port

# note: Different protocols and encryption methods correspond to different ports, when SSL/TLS is enabled, make sure the port is correct.

mailServerPort="25"

# sender

mailSender="bigdata_hjt@163.com"

# user

mailUser="bigdata_hjt@163.com"

# sender password

# note: The mail.passwd is email service authorization code, not the email login password.

mailPassword="bigdata2021"

# TLS mail protocol support

starttlsEnable="true"

# SSL mail protocol support

# only one of TLS and SSL can be in the true state.

sslEnable="false"

#note: sslTrust is the same as mailServerHost

sslTrust="smtp.163.com"

# resource storage type：HDFS,S3,NONE

resourceStorageType="HDFS"

# if resourceStorageType is HDFS，defaultFS write namenode address，HA you need to put core-site.xml and hdfs-site.xml in the conf directory.

# if S3，write S3 address，HA，for example ：s3a://dolphinscheduler，

# Note，s3 be sure to create the root directory /dolphinscheduler

defaultFS="hdfs://nameservice1" # 注意，因为这里配置了HDFS的HA

# if resourcemanager HA enable, please type the HA ips ; if resourcemanager is single, make this value empty

yarnHaIps=""

# if resourcemanager HA enable or not use resourcemanager, please skip this value setting; If resourcemanager is single, you only need to replace yarnIp1 to actual resourcemanager hostname.

singleYarnIp="data-dev01"

# resource store on HDFS/S3 path, resource file will store to this hadoop hdfs path, self configuration, please make sure the directory exists on hdfs and have read write permissions。/dolphinscheduler is recommended

resourceUploadPath="/opt/dolphinscheduler"

# who have permissions to create directory under HDFS/S3 root path

# Note: if kerberos is enabled, please config hdfsRootUser=

hdfsRootUser="root"

# kerberos config

# whether kerberos starts, if kerberos starts, following four items need to config, otherwise please ignore

kerberosStartUp="false"

# kdc krb5 config file path

krb5ConfPath="$installPath/conf/krb5.conf"

# keytab username

keytabUserName="hdfs-mycluster@ESZ.COM"

# username keytab path

keytabPath="$installPath/conf/hdfs.headless.keytab"

# api server port

apiServerPort="12345"  # 登录WebUI的初始密码

# install hosts

# Note: install the scheduled hostname list. If it is pseudo-distributed, just write a pseudo-distributed hostname

ips="10.0.2.138,10.0.2.139,10.0.2.140"

# ssh port, default 22

# Note: if ssh port is not default, modify here

sshPort="22"

# run master machine

# Note: list of hosts hostname for deploying master

masters="data-dev01"

# run worker machine

# note: need to write the worker group name of each worker, the default value is "default"

workers="data-dev01:default,data-dev02:default,data-dev03:default"

# run alert machine

# note: list of machine hostnames for deploying alert server

alertServer="data-dev03"

# run api machine

# note: list of machine hostnames for deploying api server

apiServers="data-dev01"

6、初始化

完成了上述步骤之后，确认完毕就可以进行dolphinscheduler的初始化。

sh install.sh

之后在主节点和从节点进行jps查看java进程可以看到新增的进程服务：

（主）

（从）

通过运维增加ds的映射路径：

staging-ds.***.com

通过admin账户和初始密码，可以进入到以下Web UI管理界面：

7、各目录和文件用途解释

bin：包含各种集群脚本目录

conf：配置目录

dolphinscheduler-bin：原始安装和配置目录

gc.log：记录垃圾回收日志

install.sh：初始化脚本

lib：相关jar包目录

logs：日志文件目录

pid：启动进程pid目录

script：包含各种使用脚本目录

sql：sql配置和任务记录相关目录

ui：WebUI相关目录

8、服务启停

服务的启停，都可以在./bin目录下找到对应脚本：

# 一键停止集群所有服务

sh ./bin/stop-all.sh

# 一键开启集群所有服务

sh ./bin/start-all.sh

# 启停Master

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

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

# 启停Worker

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

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

# 启停Api

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

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

# 启停Logger

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

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

# 启停Alert

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

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

9、遇到的问题和解决

1】导入基础数据失败，并报错【bin/java: No such file or directory】

解决：检查JAVA_HOME在全局环境变量下的配置正确性。

2】直接用admin在资源中心创建文件夹会提示创建资源错误NullPointerException，报空指针

解决：默认的admin租户为空首先应该创建一个租户 把admin的租户改为刚才创建的，再次创建文件夹。

10、关于dolphinscheduler的思考

查看start-all.sh脚本可以看到，内部其实包含了很多的ssh命令，从而实现跨越节点的服务启动，所以再一次证明上面的免密ssh的必要性。

另外，切换视角到dolphinscheduler-daemon.sh脚本，它里面有这样一段关于JVM配置相关的内容：

export DOLPHINSCHEDULER_OPTS="-server -Xms$HEAP_INITIAL_SIZE -Xmx$HEAP_MAX_SIZE -Xmn$HEAP_NEW_GENERATION__SIZE -XX:MetaspaceSize=128m -XX:MaxMetaspaceSize=128m  -Xss512k -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:+CMSParallelRemarkEnabled -XX:LargePageSizeInBytes=128m -XX:+UseCMSInitiatingOccupancyOnly -XX:CMSInitiatingOccupancyFraction=70 -XX:+PrintGCDetails -Xloggc:gc.log -XX:+HeapDumpOnOutOfMemoryError  -XX:HeapDumpPath=dump.hprof"

设置的MaxMetaspaceSize=128m，这里在后续的元数据增加场景可以增大，-XX:+UseConcMarkSweepGC使用的CMS垃圾回收器，其实可以修改为G1垃圾回收器，但是它里面增加一些关于CMS的配置，修改为G1对应的配置即可。

另外-XX+UseCMSInitiatingOccupancyOnly标志来命令JVM不基于运行时收集的数据来启动CMS垃圾收集周期，和CMSInitiatingOccupancyFraction=70，说明在对CMS GC频率上，是做了一定优化的。这个参数的意思是：设定CMS在对内存占用率达到70%的时候开始GC(因为CMS会有浮动垃圾,所以一般都较早启动GC)。

关于这一点，确实能发现对比Oozie的优化做的还是不错的，因为一般来说，JVM自己就能做非常好的垃圾回收策略，在apache基金会下面的孵化项目中，dolphinscheduler敢于这样做出设置，都是在application产生的对象的生命周期有深刻的认知之后才会使用这个标志。并且因为涉及到Haoop、Spark、Hive等多个组件，能兼容管理，确实也说明了这个组件的可取之处的。

以上，为海豚调度器本次的安装实录。