企业级大数据安全架构（十一）Kerberos接入dophinscheduler

作者：楼高建议将dophinscheduler集成到Ambari安装部署，在Ambari上面开启kerberos

1.安装准备

编译从GitHub获取dolphinscheduler-1.3.9源码

git clone https://github.com/apache/dolphinscheduler.git -b 1.3.9-release

https://github.com/apache/dolphinscheduler/archive/2.0.3.zip进入源码目录进行编译

mvn -U clean install -Prpmbuild -Dmaven.test.skip=true -X

编译完成后在dolphinscheduler-dist/target/rpm/apache-dolphinscheduler/RPMS/noarch路径下可以找到打包好的rpm将编译好的包发送到其他节点

注意：nodejs版本不要太低，不然编译过程不容易通过，另外npm尽量不要使用淘宝镜像

2.环境准备2.1创建部署用户

选择需要部署dolphinscheduler的机器创建系统用户dolphinscheduler

创建部署用户，并且一定要配置sudo免密。以创建dolphinscheduler用户为例

useradd dolphinscheduler;echo "dolphinscheduler" | passwd --stdin dolphinscheduler # 配置sudo免密echo 'dolphinscheduler ALL=(ALL) NOPASSWD: NOPASSWD: ALL' >> /etc/sudoerssed -i 's/Defaults requirett/#Defaults requirett/g' /etc/sudoers # 修改目录权限，使得部署用户对dolphinscheduler-backend目录有操作权限 chown -R dolphinscheduler:dolphinscheduler dolphinscheduler2.2在部署用户并配置 ssh 本机免密登录su dolphinscheduler;ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsacat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keyschmod 600 ~/.ssh/authorized_keys2.3 其他待部署的机器配置ssh免密su dolphinscheduler;for ip in db003.hdp.com db004.hdp.com; do ssh-copy-id $ip # 该操作执行过程中需要手动输入 dolphinscheduler 用户的密码done2.4数据库初始化

（1）登陆数据库：mysql -uroot -phdp123456

（2）创建数据库并设置相应权限

CREATE DATABASE dolphinscheduler;GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'dolphinscheduler'@'%' IDENTIFIED BY 'hdp123456';GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'dolphinscheduler'@'localhost' IDENTIFIED BY 'hdp123456';GRANT ALL PRIVILEGES ON dolphinscheduler.* TO 'dolphinscheduler'@'current_hostname' IDENTIFIED BY 'hdp123456';FLUSH PRIVILEGES;2.5选择需要部署dolphinscheduler的机器进行rpm安装

执行安装命令rpm -ivh apache-dolphinscheduler-1.3.10-1.noarch.rpm

安装完成后将mysql驱动包放到dolphinscheduler/lib目录下

2.6配置调度所执行任务的必要环境

修改dolphinscheduler目录conf下的dolphinscheduler_env.sh 环境变量

export HADOOP_HOME=/usr/hdp/3.1.0.0-78/hadoopexport HADOOP_CONF_DIR=/usr/hdp/3.1.0.0-78/hadoop/etc/hadoop/export SPARK_HOME2=/usr/hdp/3.1.0.0-78/spark2export PYTHON_HOME=/opt/soft/pythonexport JAVA_HOME=/opt/module/jdk1.8.0_281export HIVE_HOME=/usr/hdp/3.1.0.0-78/hiveexport FLINK_HOME=/usr/hdp/3.1.0.0-78/flinkexport PATH=$HADOOP_HOME/bin:$SPARK_HOME2/bin:$PYTHON_HOME:$JAVA_HOME/bin:$HIVE_HOME/bin:$PATH:$FLINK_HOME/bin:$PATH

组件配置路径根据自己安装路径配置，但是jdk必须软链到/usr/bin/java，否则后续安装完成之后启动会报错java命令找不到

ln -s /opt/module/jdk1.8.0_281/bin/java /usr/bin/java

3.Ambari集成dophinscheduler3.1部署ambari插#将dolphinscheduler自定义集成配置文件拷贝到ambari-server服务节点的相关目录cp -r ambari_plugin/common-services/DOLPHIN /var/lib/ambari-server/resources/common-services/cp -r ambari_plugin/statcks/DOLPHIN /var/lib/ambari-server/resources/stacks/HDP/3.1/services/3.2重启ambari-server

在ambari server部署节点执行重启命令ambari-server restart

3.3开始安装

（1）进入ambari界面，界面左侧 >> services >> Add service >> Dolphin Scheduler >> NEXT >>选择DS Master