如何为oozie工作流程中的不同操作指定不同的用户库路径
我有一个火花动作和一个Java动作
<workflow-app name="batch-publisher" xmlns="uri:oozie:workflow:0.5">
<start to="spark-a5b5"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="spark-a5b5">
<spark xmlns="uri:oozie:spark-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.libpath</name>
<value>${nameNode}/user/hdfs/sps-batch-bi/Jars/</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
</configuration>
<master>yarn</master>
<mode>client</mode>
<name>MySpark2</name>
<class>com.sps.analytics.batch.bi.BatchApp</class>
<jar>${nameNode}/user/hdfs/sps-batch-bi/Jars/sps-analytics-batch-bi-assembly-0.1.jar</jar>
<spark-opts>--driver-memory 4G --executor-memory 4G --num-executors 2</spark-opts>
<arg>/user/hdfs/sps-batch-bi/config.yaml</arg>
<arg>/user/hdfs/sps-batch-bi/Query_sps_txn_Detail.xlsx</arg>
<arg>spark</arg>
</spark>
<ok to="java-8283"/>
<error to="Kill"/>
</action>
<action name="java-8283">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>oozie.launcher.mapreduce.map.env</name>
<value>LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/tableau/lib64/tableausdk/</value>
</property>
<property>
<name>oozie.launcher.mapreduce.user.classpath.first</name>
<value>true</value>
</property>
</configuration>
<main-class>com.sps.analytics.bi.publisher.PublisherApp</main-class>
<arg>sps-data-bi-dev</arg>
<arg>default</arg>
<arg>default</arg>
<file>/user/sps/bi-publisher/config/application.properties#application.properties</file>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>
如何为两个动作指定不同的库路径。我在这两个装配罐中都有罐子。
答案 0 :(得分:0)
进行操作的子工作流程有助于解决jar混乱的问题
答案 1 :(得分:0)
你可以试试这个
<action name="java-8283">
...
<java>
<file>hdfs://xx/xx/xx.jar</file>
<file>hdfs://xx/xx/xx.jar</file>
</java>
<ok to="End"/>
...
</action>