DolpinScheduler2.0.5详细介绍与踩坑实战

本文涉及的产品
RDS MySQL Serverless 基础系列,0.5-2RCU 50GB
云数据库 RDS MySQL,集群系列 2核4GB
推荐场景:
搭建个人博客
RDS MySQL Serverless 高可用系列,价值2615元额度,1个月
简介: 大数据之DolpinScheduler2.0.5详细介绍与踩坑实战

1 简介

官网:https://dolphinscheduler.apache.org/zh-cn/

文档:https://dolphinscheduler.apache.org/zh-cn/docs/

https://dolphinscheduler.apache.org/zh-cn/docs/2.0.5/user_doc/guide/flink-call.html

Apache DolphinScheduler是一个分布式 易扩展可视化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关系,使调度系统在数据处理流程中开箱即用。

版本:2.0.5

1.1 特性

  • 简单易用:DAG监控界面,所有流程定义都是可视化,通过拖拽任务完成定制DAG,通过API方式与第三方系统集成,一键部署。
  • 高可靠性:去中心化的多Master和多Worker服务对等架构,避免单Master压力过大,另外采用任务缓冲队列来避免过载。
  • 高扩展性:支持自定义任务类型,调度器使用分布式调度,调度能力随集群线性增长,Master和Worker支持动态上下线。
  • 丰富的使用场景:支持多租户,支持暂停恢复操作,紧密贴合大数据生态,提供Spark、Hive、MapRedece、Python、Sub_process、Shell等近20中任务类型。

1.2 核心架构(主要角色)

  • MasterServer:采用分布式务中心设计理念,MasterServer主要负责DAG任务切分,任务提交、任务监控,并同时监听其他MasterServer和WorkerServer的健康状态。(负责工作流)
  • WorkerServer:也采用分布式务中心设计理念,WorkerServer主要负责任务的执行和提供日志服务。(负责任务)
  • ZooKeeper:系统中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。

在这里插入图片描述

2 部署说明

2.1 软硬件要求

操作系统要求

操作系统 版本
Red Hat Enterprise Linux >=7.0
CentOS >=7.0
Oracle Enterprise Linux >=7.0
Ubuntu LTS >=16.04

服务器硬件要求

CPU 内存 硬盘类型 网络 实例数量
4核+ 8GB+ SAS 千兆网卡 1+

2.2 部署模式

2.2.1 单机模式

单机模式(sandalone)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,即可一键启动DolphinScheduler。(学习、测试)

2.2.2 伪集群模式

伪集群模式(Pseudo-Cluster)是单台机器部署DolphinSchdeuler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和数据库需要单独安装并进行响应配置。

2.2.3 集群模式

相对于伪集群,区别在于可以在多台机器部署 DolphinScheduler各项服务,并且Maser、Worker等服务可配置多个。

3 集群搭建

下载:https://dlcdn.apache.org/dolphinscheduler/2.0.5/

3.1 环境准备

3.1.1 硬件

Master:通常2~3个
Worker:多个

这里配置一个Master、三个Worker

名称 IP 角色
master 192.168.56.20 master、worker
slave1 192.168.56.21 worker
slave2 192.168.56.22 worker

3.1.2 软件

  1. 三台阶段均需安装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树分析工具psmisc

    sudo yum install -y psmisc

本次环境:
JAVA:V1.8.0_151
ZooKeeper:V3.7.1
驱动:
mysql-connector-java-8.0.16.jar

3.2 创建元数据库及用户

安装docker:https://blog.csdn.net/zx77588023/article/details/122652560

docker run -d \
-p 3306:3306 \
-v /usr/local/mysql/data/conf:/etc/mysql/conf.d \
-v /usr/local/mysql/data/data:/var/lib/mysql \
-e MYSQL_ROOT_PASSWORD=root \
--name ds_mysql mysql:5.7

登录mysql

mysql -u root -p

1、创建数据库

CREATE DATABASE dolphinscheduler DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;

2、创建用户

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

如果测试使用,提供密码过于简单,可以降低密码强度级别

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

3、赋予用户响应权限

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

3.3 初始化数据库

1、拷贝MySQL驱动到DolphinScheduler的解压目录lib中,要求版本:JDBC Driver 8.0.16

cp mysql-connector-java-8.0.16.jar /usr/local/ds/apache-dolphinscheduler-2.0.5-bin/lib

2、执行数据库初始化脚本

cd /usr/local/ds/apache-dolphinscheduler-2.0.5-bin
./script/create-dolphinscheduler.sh

3.4 配置一键部署脚本

三台机器均配置免密超级用户 charles

useradd charles -u 520
sudo vim /etc/sudoers
charles ALL=(ALL)       NOPASSWD:ALL
# 创建用户需使用 root 登录
useradd dolphinscheduler

# 添加密码
echo "dolphinscheduler" | passwd --stdin dolphinscheduler

# 配置 sudo 免密
sed -i '$adolphinscheduler  ALL=(ALL)  NOPASSWD: NOPASSWD: ALL' /etc/sudoers
sed -i 's/Defaults    requirett/#Defaults    requirett/g' /etc/sudoers

# 修改目录权限,使得部署用户对二进制包解压后的 apache-dolphinscheduler-*-bin 目录有操作权限
chown -R dolphinscheduler:dolphinscheduler apache-dolphinscheduler-*-bin

修改 conf/config目录下得install_config.conf文件

vim conf/config/install_config.conf
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License.  You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#

# ---------------------------------------------------------
# INSTALL MACHINE
# ---------------------------------------------------------
# A comma separated list of machine hostname or IP would be installed DolphinScheduler,
# including master, worker, api, alert. If you want to deploy in pseudo-distributed
# mode, just write a pseudo-distributed hostname
# Example for hostnames: ips="ds1,ds2,ds3,ds4,ds5", Example for IPs: ips="192.168.8.1,192.168.8.2,192.168.8.3,192.168.8.4,192.168.8.5"
# 将要部署的任一 DolpinScheduler 服务的主机名或ip列表
ips="master,slave1,slave2"

# Port of SSH protocol, default value is 22. For now we only support same port in all `ips` machine
# modify it if you use different ssh port
sshPort="22"

# A comma separated list of machine hostname or IP would be installed Master server, it
# must be a subset of configuration `ips`.
# Example for hostnames: masters="ds1,ds2", Example for IPs: masters="192.168.8.1,192.168.8.2"
# master所在主机名列表,必须是 ips 的子集
masters="master"

# A comma separated list of machine <hostname>:<workerGroup> or <IP>:<workerGroup>.All hostname or IP must be a
# subset of configuration `ips`, And workerGroup have default value as `default`, but we recommend you declare behind the hosts
# Example for hostnames: workers="ds1:default,ds2:default,ds3:default", Example for IPs: workers="192.168.8.1:default,192.168.8.2:default,192.168.8.3:default"
# work 主机名及队列,此处的ip 必须在 ips 列表中
workers="master:default,slave1:default,slave2:default"

# A comma separated list of machine hostname or IP would be installed Alert server, it
# must be a subset of configuration `ips`.
# Example for hostname: alertServer="ds3", Example for IP: alertServer="192.168.8.3"
# 告警服务所在服务器主机名
alertServer="slave1"

# A comma separated list of machine hostname or IP would be installed API server, it
# must be a subset of configuration `ips`.
# Example for hostname: apiServers="ds1", Example for IP: apiServers="192.168.8.1"
# api 服务所在服务器主机名
apiServers="slave1"

# A comma separated list of machine hostname or IP would be installed Python gateway server, it
# must be a subset of configuration `ips`.
# Example for hostname: pythonGatewayServers="ds1", Example for IP: pythonGatewayServers="192.168.8.1"

# 不需要的配置项,可以保留默认值,也可以 #注释
#pythonGatewayServers="ds1"

# The directory to install DolphinScheduler for all machine we config above. It will automatically be created by `install.sh` script if not exists.
# Do not set this configuration same as the current path (pwd)
# 安装路径
installPath="/usr/local/ds/ds_install"

# The user to deploy DolphinScheduler for all machine we config above. For now user must create by yourself before running `install.sh`
# script. The user needs to have sudo privileges and permissions to operate hdfs. If hdfs is enabled than the root directory needs
# to be created by this user

# 部署用户 具有sudo权限,配置免密 
deployUser="charles"
#deployUser="root"

# The directory to store local data for all machine we config above. Make sure user `deployUser` have permissions to read and write this directory.

# 临时路径
dataBasedirPath="/tmp/dolphinscheduler"

# ---------------------------------------------------------
# DolphinScheduler ENV
# ---------------------------------------------------------
# JAVA_HOME, we recommend use same JAVA_HOME in all machine you going to install DolphinScheduler
# and this configuration only support one parameter so far.
javaHome="/usr/local/jdk1.8.0_151"

# DolphinScheduler API service port, also this is your DolphinScheduler UI component's URL port, default value is 12345

apiServerPort="12345"

# ---------------------------------------------------------
# Database
# NOTICE: If database value has special characters, such as `.*[]^${}\+?|()@#&`, Please add prefix `\` for escaping.
# ---------------------------------------------------------
# The type for the metadata database
# Supported values: ``postgresql``, ``mysql`, `h2``.

# 数据库类型
DATABASE_TYPE="mysql"

# string jdbc:mysql://127.0.0.1:3306/dolphinscheduler?useUnicode=true&characterEncoding=UTF-8 as example
SPRING_DATASOURCE_URL="jdbc:mysql://master:3306/dolphinscheduler?useUnicode=true&characterEncoding=UTF-8"

# 数据库用户名
SPRING_DATASOURCE_USERNAME="dolphinscheduler"

# 数据库密码
SPRING_DATASOURCE_PASSWORD="dolphinscheduler"

# ---------------------------------------------------------
# Registry Server(这里配置zookeeper)
# ---------------------------------------------------------
# Registry Server plugin name, should be a substring of `registryPluginDir`, DolphinScheduler use this for verifying configuration consistency
registryPluginName="zookeeper"

# Registry Server address.
registryServers="master:2181,slave1:2181,slave2:2181"

# Registry Namespace(注册中心命名空间)
registryNamespace="dolphinscheduler"

# ---------------------------------------------------------
# Worker Task Server
# ---------------------------------------------------------
# Worker Task Server plugin dir. DolphinScheduler will find and load the worker task plugin jar package from this dir.
taskPluginDir="lib/plugin/task"

# resource storage type: HDFS, S3, NONE(资源中心,比存储MR中使用的jar包)
resourceStorageType="HDFS"  # 这里我们存储在HDFS中

# resource store on HDFS/S3 path, resource file will store to this hdfs path, self configuration, please make sure the directory exists on hdfs and has read write permissions. "/dolphinscheduler" is recommended
# 存储资源的根目录
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
# 默认FS得知(我们用hadoop的)
defaultFS="hdfs://master:9000"

# if resourceStorageType is S3, the following three configuration is required, otherwise please ignore
s3Endpoint="http://192.168.xx.xx:9010"
s3AccessKey="xxxxxxxxxx"
s3SecretKey="xxxxxxxxxx"

# 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 node, keep this value empty
# 启用了HA
yarnHaIps=""

# if resourcemanager HA is enabled or not use resourcemanager, please keep the default value; If resourcemanager is single node, you only need to replace 'yarnIp1' to actual resourcemanager hostname
# 未启用HA
singleYarnIp="master"

# who has permission to create directory under HDFS/S3 root path
# Note: if kerberos is enabled, please config hdfsRootUser=
hdfsRootUser="hdfs"

# kerberos config
# whether kerberos starts, if kerberos starts, following four items need to config, otherwise please ignore
# hdfs超级用户(谁启动的)
kerberosStartUp="false"
# kdc krb5 config file path
krb5ConfPath="$installPath/conf/krb5.conf"
# keytab username,watch out the @ sign should followd by \\
keytabUserName="hdfs-mycluster\\@ESZ.COM"
# username keytab path
keytabPath="$installPath/conf/hdfs.headless.keytab"
# kerberos expire time, the unit is hour
kerberosExpireTime="2"

# use sudo or not
sudoEnable="true"

# worker tenant auto create
workerTenantAutoCreate="true"

3.5 一键部署 DolpinScheduler

1、启动 Zookeeper 集群

zh.sh start

2、一键部署并启动DolphinScheduler

./install.sh

3、查看 DolphinScheduler进程

jps

4、访问DophinScheduler UI

http://192.168.56.21:12345/dolphinscheduler

初始用户名:admin
初始密码:dolphinscheduler123

在这里插入图片描述

3.6 DolphinScheduler启停命令

进入到安装目录 cd /usr/local/ds/ds_install
1、一键启停所有服务

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

2、启停 Master

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

3、启停 Worker

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

4、启停 Api

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

5、启停 Logger

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

4 DolphinScheduler入门

4.1 安全中心配置

4.1.1 Yarn队列管理

队列是在执行spark、mapreduce等程序,需要用到“队列”参数时使用的。
此处的队列对应的是Yarn调度器资源队列。故队列概念只对泡在Yarn上的任务类型有效。此处创建出的队列,可供后续任务进行选择。需要注意的是,在DolphinSheduler中创建队列,并不会影响到Yarn调度器的队列配置。

4.1.2 租户管理

租户对应的是 Linux的用户,用于worker提交作业所使用的用户。如果 linux 没有这个用户,默认情况下回导致任务运行失败。可以通过修改 worker.properties 配置文件中参数 worker.tenant.auto.create=true(默认值为flase)实现当linux用户不存在时自动创建该用户。
worker.tenant.auto.create=true 参数会要求 worker 可以免密运行 sudo 命令。

vim /usr/local/ds/ds_install/conf/worker.properties

报错处理:

org.apache.dolphinscheduler.registry.api.RegistryException: zookeeper connect timeout

vim registry.properties
registry.connection.timeout.ms=75000

org.apache.dolphinscheduler.api.exceptions.ApiExceptionHandler:[46] - 创建租户错误
查看

vim /usr/local/hadoop-2.8.4/etc/hadoop/core-site.xml
<!--配置hdfs默认的命名-->
<property>
    <name>fs.defaultFS</name>
    <value>hdfs://master:9000</value>
</property>

资源存储类型:HDFS,S3,NONE,修改 install_config.conf

如果resourceStorageType是HDFS,defaultFS写入namenode地址,HA需要把core-site.xml和hdfs-site.xml放到conf目录下。
如果是S3,写S3地址,HA,例如:s3a://dolphinscheduler,
注意,s3一定要创建根目录/dolphinscheduler

org.apache.hadoop.ipc.RemoteException: Permission denied: user=hdfs, access=WRITE, inode="/":root:supergroup:drwxr-xr-x

org.apache.hadoop.security.AccessControlException: Permission denied: user=hdfs, access=WRITE, inode="/":root:supergroup:drwxr-xr-x
对非CDH用户

vim /usr/local/hadoop-2.8.4/etc/hadoop/hdfs-site.xml
<!--是否启用hdfs的权限-->
<property>
    <name>dfs.permissions</name>
    <value>false</value>
</property>

参考地址:
https://blog.csdn.net/g11d111/article/details/72902112

参考地址:
https://www.bilibili.com/video/BV1sa411x7Ep
https://github.com/apache/dolphinscheduler/issues/4389

相关实践学习
基于MSE实现微服务的全链路灰度
通过本场景的实验操作,您将了解并实现在线业务的微服务全链路灰度能力。
相关文章
|
存储 移动开发 JSON
前端面试题汇总大全 -- 持续更新!(一)
前端面试题汇总大全 -- 持续更新!
446 0
|
7月前
|
存储 前端开发 JavaScript
前端面试题整理 (ES6篇)
前端面试题整理 (ES6篇)
|
存储 安全 Shell
极客之命令注入实操第二篇
极客之命令注入实操第二篇
|
消息中间件 SQL 缓存
Java面试通关:阿里内部实战模拟面试精讲题库,竟被上传GitHub!
最近收到小伙伴的反馈: 事实也确实是这样,如今不像从前,会个CRUD就可以随随便便找个10几k的工作,现在面试中级的话都从底层问到框架,从jvm,juc底层问到消息队列高并发了,还问SQL的调优。至于高级都是对于消息队列、缓存、分布式等各种主流技术还需要知道原理! 其实这些技术在真正的实际开发环境中真的用得到吗?不一定的,但是没办法,环境就是如此,能做的也就只有不断提升自己,去适应市场环境,提高自身技术水平!
|
移动开发 编解码 前端开发
前端面试题汇总大全 -- 持续更新!(三)
前端面试题汇总大全 -- 持续更新!
95 0
|
存储 缓存 JSON
前端面试题汇总大全 -- 持续更新!(四)
前端面试题汇总大全 -- 持续更新!
112 0
|
存储 缓存 JavaScript
前端面试题汇总大全 -- 持续更新!(二)
前端面试题汇总大全 -- 持续更新!
95 0
|
编译器 C语言 C++
【C++系列P1】带上这篇基础小宝典,进发C++!(持续更新ing~)
【C++系列P1】带上这篇基础小宝典,进发C++!(持续更新ing~)
|
前端开发 JavaScript ice
flexiwan项目踩坑实践
flexiManage是以色列一家初创公司flexiWAN开源的基于SD-WAN平台的应用层的框架,包括flexiManage服务端框架以及硬件侧的flexiAgent框架,然而其并没有开源前端框架,为了验证其SD-WAN方案的可行性,需要快速搭建一个前端应用
290 0
|
Java 数据库
开发踩坑之判等问题
开发过程中避免不了判等问题?你在判等的过程中有没有遇到过坑呢?那么你是怎么解决的?快来看看我们是不是一样。
113 1
开发踩坑之判等问题

相关实验场景

更多