114 lines
3.9 KiB
XML
114 lines
3.9 KiB
XML
<workflow-app name="hosted_by_map" xmlns="uri:oozie:workflow:0.5">
|
|
|
|
<parameters>
|
|
<property>
|
|
<name>sourcePath</name>
|
|
<description>the source path</description>
|
|
</property>
|
|
<property>
|
|
<name>outputPath</name>
|
|
<description>the output path</description>
|
|
</property>
|
|
<property>
|
|
<name>sparkDriverMemory</name>
|
|
<description>memory for driver process</description>
|
|
</property>
|
|
<property>
|
|
<name>sparkExecutorMemory</name>
|
|
<description>memory for individual executor</description>
|
|
</property>
|
|
<property>
|
|
<name>sparkExecutorCores</name>
|
|
<description>number of cores used by single executor</description>
|
|
</property>
|
|
<property>
|
|
<name>oozieActionShareLibForSpark2</name>
|
|
<description>oozie action sharelib for spark 2.*</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2ExtraListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorAppListener</value>
|
|
<description>spark 2.* extra listeners classname</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2SqlQueryExecutionListeners</name>
|
|
<value>com.cloudera.spark.lineage.NavigatorQueryListener</value>
|
|
<description>spark 2.* sql query execution listeners classname</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2YarnHistoryServerAddress</name>
|
|
<description>spark 2.* yarn history server address</description>
|
|
</property>
|
|
<property>
|
|
<name>spark2EventLogDir</name>
|
|
<description>spark 2.* event log dir location</description>
|
|
</property>
|
|
</parameters>
|
|
|
|
<global>
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<configuration>
|
|
<property>
|
|
<name>mapreduce.job.queuename</name>
|
|
<value>${queueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.launcher.mapred.job.queue.name</name>
|
|
<value>${oozieLauncherQueueName}</value>
|
|
</property>
|
|
<property>
|
|
<name>oozie.action.sharelib.for.spark</name>
|
|
<value>${oozieActionShareLibForSpark2}</value>
|
|
</property>
|
|
|
|
</configuration>
|
|
</global>
|
|
|
|
<start to="downloadDOAJ"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
|
|
<action name="downloadDOAJ">
|
|
<shell xmlns="uri:oozie:shell-action:0.2">
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<configuration>
|
|
<property>
|
|
<name>mapred.job.queue.name</name>
|
|
<value>${queueName}</value>
|
|
</property>
|
|
</configuration>
|
|
<exec>download.sh</exec>
|
|
<argument>${doajJsonFileURL}</argument>
|
|
<argument>${dumpPath}</argument>
|
|
<argument>${dumpFileName}</argument>
|
|
<env-var>HADOOP_USER_NAME=${wf:user()}</env-var>
|
|
<file>download.sh</file>
|
|
<capture-output/>
|
|
</shell>
|
|
<ok to="extractTarGzAndMap"/>
|
|
<error to="Kill"/>
|
|
|
|
</action>
|
|
|
|
<action name="extractTarGzAndMap">
|
|
<java>
|
|
<main-class>eu.dnetlib.dhp.oa.graph.hostedbymap.ExtractAndMapDoajJson</main-class>
|
|
<arg>--hdfsNameNode</arg><arg>${nameNode}</arg>
|
|
<arg>--compressedFile</arg><arg>${dumpPath}/${dumpFileName}</arg>
|
|
<arg>--workingPath</arg><arg>${workingDir}/DOAJ/</arg>
|
|
<arg>--outputFile</arg><arg>/user/${wf:user()}/data/doaj.json</arg>
|
|
</java>
|
|
<ok to="produceHBM"/>
|
|
<error to="End"/>
|
|
</action>
|
|
|
|
|
|
|
|
<end name="End"/>
|
|
|
|
</workflow-app> |