dnet-hadoop/dhp-workflows/dhp-continuous-validation/src/main/resources/eu/dnetlib/dhp/continuous_validator/oozie_app/workflow.xml

111 lines
4.7 KiB
XML

<workflow-app name="Validate metadata records against OpenAIRE Guidelines" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>parquet_file_path</name>
<value>./src/test/resources/part-00589-733117df-3822-4fce-bded-17289cc5959a-c000.snappy.parquet</value>
<description>the full path of the parquet file</description>
</property>
<property>
<name>openaire_guidelines</name>
<value>4.0</value>
<description>the version of the OpenAIRE Guidelines to validate the records against</description>
</property>
<property>
<name>outputPath</name>
<value>.</value>
<description>the path of the output-directory where the result-json-files will be stored</description>
</property>
<property>
<name>sparkDriverMemory</name>
<value>4096M</value>
<description>memory for driver process</description>
</property>
<property>
<name>sparkExecutorMemory</name>
<value>2048m</value>
<description>memory for individual executor</description>
</property>
<property>
<name>sparkExecutorCores</name>
<value>4</value>
<description>number of cores used by single executor</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>oozieActionShareLibForSpark2</name>
<description>oozie action sharelib for spark 2.*</description>
</property>
<property>
<name>spark2YarnHistoryServerAddress</name>
<description>spark 2.* yarn history server address</description>
</property>-->
<property>
<name>spark2EventLogDir</name>
<value>spark2/logs</value>
<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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="ValidateMetadataRecords"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ValidateMetadataRecords">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Validate multiple records against OpenAIRE Guidelines</name>
<class>eu.dnetlib.dhp.continuous_validator.ContinuousValidator</class>
<jar>dhp-continuous-validation-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--conf spark.executor.memoryOverhead=${sparkExecutorMemoryOverhead}
--executor-cores=${sparkExecutorCores}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<!-- Arguments passed to the "main" method of the class defined above. -->
<arg>--parquet_file_path</arg><arg>${parquet_file_path}</arg>
<arg>--openaire_guidelines</arg><arg>${openaire_guidelines}</arg>
<arg>--outputPath</arg><arg>${outputPath}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>