一、 DolphinScheduler简介

1.1 DolphinScheduler概述

  Apache DolphinScheduler是一个分布式、易扩展的可视化DAG工作流任务调度平台。致力于解决数据处理流程中错综复杂的依赖关系，使调度系统在数据处理流程中开箱即用。

1.2 DolphinScheduler核心架构

DolphinScheduler的主要角色如下：

• MasterServer 采用分布式无中心设计理念，MasterServer主要负责 DAG 任务切分、任务提交、任务监控，并同时监听其它MasterServer和WorkerServer的健康状态。
• WorkerServer 也采用分布式无中心设计理念，WorkerServer主要负责任务的执行和提供日志服务。
• ZooKeeper服务，系统中的MasterServer和WorkerServer节点都通过ZooKeeper来进行集群管理和容错。
• Alert服务，提供告警相关服务。
• API接口层，主要负责处理前端UI层的请求。
• UI，系统的前端页面，提供系统的各种可视化操作界面。
  

二、DolphinScheduler部署说明

2.1 软硬件环境要求

2.1.1 操作系统版本要求

2.1.2 服务器硬件要求

2.2 部署模式

DolphinScheduler支持多种部署模式，包括单机模式（Standalone）、伪集群模式（Pseudo-Cluster）、集群模式（Cluster）等。

2.2.1 单机模式

单机模式（standalone）模式下，所有服务均集中于一个StandaloneServer进程中，并且其中内置了注册中心Zookeeper和数据库H2。只需配置JDK环境，就可一键启动DolphinScheduler，快速体验其功能。

2.2.2 伪集群模式

伪集群模式（Pseudo-Cluster）是在单台机器部署 DolphinScheduler 各项服务，该模式下master、worker、api server、logger server等服务都只在同一台机器上。Zookeeper和数据库需单独安装并进行相应配置。

2.2.3 集群模式

集群模式（Cluster）与伪集群模式的区别就是在多台机器部署 DolphinScheduler各项服务，并且可以配置多个Master及多个Worker。

三、DolphinScheduler集群模式部署

3.1 集群规划

集群模式下，可配置多个Master及多个Worker。通常可配置2~3个Master，若干个Worker。由于集群资源有限，此处配置一个Master，三个Worker，集群规划如下。

3.2 前置准备工作

1）三台节点均需部署JDK（1.8+），并配置相关环境变量。
2）需部署数据库，支持MySQL（5.7+）或者PostgreSQL（8.2.15+）。
3）需部署Zookeeper（3.4.6+）。
4）三台节点均需安装进程管理工具包psmisc。

[atguigu@hadoop102 ~]$ sudo yum install -y psmisc
[atguigu@hadoop103 ~]$ sudo yum install -y psmisc
[atguigu@hadoop104 ~]$ sudo yum install -y psmisc

3.3 解压DolphinScheduler安装包

1）上传DolphinScheduler安装包到hadoop102节点的/opt/software目录
2）解压安装包到当前目录

[atguigu@hadoop102 software]$ tar -zxvf apache-dolphinscheduler-1.3.9-bin.tar.gz 

注：解压目录并非最终的安装目录

3.4 初始化数据库

DolphinScheduler 元数据存储在关系型数据库中，故需创建相应的数据库和用户。
1）创建数据库

mysql> CREATE DATABASE dolphinscheduler DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;

2）创建用户

mysql> CREATE USER 'dolphinscheduler'@'%' IDENTIFIED BY 'dolphinscheduler';

注：
若出现以下错误信息，表明新建用户的密码过于简单。
ERROR 1819 (HY000): Your password does not satisfy the current policy requirements
可提高密码复杂度或者执行以下命令降低MySQL密码强度级别。

mysql> set global validate_password_length=4;
mysql> set global validate_password_policy=0;

3）赋予用户相应权限

mysql> GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'dolphinscheduler'@'%';
mysql> flush privileges;

4）修改数据源配置文件
进入DolphinScheduler解压目录

cd /opt/software/apache-dolphinscheduler-1.3.9-bin/

修改conf目录下的datasource.properties文件

vim conf/datasource.properties

修改内容如下

spring.datasource.driver-class-name=com.mysql.jdbc.Driver
spring.datasource.url=jdbc:mysql://hadoop102:3306/dolphinscheduler?useUnicode=true&characterEncoding=UTF-8
spring.datasource.username=dolphinscheduler
spring.datasource.password=dolphinscheduler

5）拷贝MySQL驱动到DolphinScheduler的解压目录下的lib中

cp /opt/software/mysql-connector-java-5.1.27-bin.jar lib/

6）执行数据库初始化脚本
数据库初始化脚本位于DolphinScheduler解压目录下的script目录中，即/opt/software/ds/apache-dolphinscheduler-1.3.9-bin/script/。

script/create-dolphinscheduler.sh

3.5 配置一键部署脚本

修改解压目录下的conf/config目录下的install_config.conf文件

 vim conf/config/install_config.conf 

修改内容如下

# postgresql or mysql
dbtype="mysql"

# db config
# db address and port
dbhost="hadoop102:3306"

# 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="dolphinscheduler"

# zk cluster
zkQuorum="hadoop102:2181,hadoop103:2181,hadoop104:2181"

# Note: the target installation path for dolphinscheduler, please not config as the same as the current path (pwd)
installPath="/opt/module/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="atguigu"

# resource storage type: HDFS, S3, NONE
resourceStorageType="HDFS"

# 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="/dolphinscheduler"

# 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://hadoop102:8020"

# resourcemanager port, the default value is 8088 if not specified
resourceManagerHttpAddressPort="8088"

# if resourcemanager HA is enabled, please set the HA IPs; if resourcemanager is single, keep this value empty
yarnHaIps=

# if resourcemanager HA is enabled or not use resourcemanager, please keep the default value; If resourcemanager is single, you only need to replace ds1 to actual resourcemanager hostname
singleYarnIp="hadoop103"

# who have permissions to create directory under HDFS/S3 root path
# Note: if kerberos is enabled, please config hdfsRootUser=
hdfsRootUser="atguigu"

# api server port
apiServerPort="12345"


# install hosts
# Note: install the scheduled hostname list. If it is pseudo-distributed, just write a pseudo-distributed hostname
ips="hadoop102,hadoop103,hadoop104"

# 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="hadoop102"

# run worker machine
# note: need to write the worker group name of each worker, the default value is "default"
workers="hadoop102:default,hadoop103:default,hadoop104:default"

# run alert machine
# note: list of machine hostnames for deploying alert server
alertServer="hadoop102"

# run api machine
# note: list of machine hostnames for deploying api server
apiServers="hadoop102"

3.6 一键部署DolphinScheduler

1）启动Zookeeper集群

zk.sh start

2）一键部署并启动DolphinScheduler

./install.sh 

3）查看DolphinScheduler进程

--------- hadoop102 ----------
29139 ApiApplicationServer
28963 WorkerServer
3332 QuorumPeerMain
2100 DataNode
28902 MasterServer
29081 AlertServer
1978 NameNode
29018 LoggerServer
2493 NodeManager
29551 Jps
--------- hadoop103 ----------
29568 Jps
29315 WorkerServer
2149 NodeManager
1977 ResourceManager
2969 QuorumPeerMain
29372 LoggerServer
1903 DataNode
--------- hadoop104 ----------
1905 SecondaryNameNode
27074 WorkerServer
2050 NodeManager
2630 QuorumPeerMain
1817 DataNode
27354 Jps
27133 LoggerServer

4）访问DolphinScheduler UI
DolphinScheduler UI地址为http://hadoop102:12345/dolphinscheduler
初始用户的用户名为：admin，密码为dolphinscheduler123

3.7 DolphinScheduler启停命令

DolphinScheduler的启停脚本均位于其安装目录的bin目录下。
1）一键启停所有服务

./bin/start-all.sh
./bin/stop-all.sh

注意同Hadoop的启停脚本进行区分。

2）启停 Master

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

3）启停 Worker

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

4）启停 Api

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

5）启停 Logger

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

6）启停 Alert

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