forked from D-Net/dnet-hadoop
98 lines
4.0 KiB
XML
98 lines
4.0 KiB
XML
<workflow-app name="blacklisting" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>postgresURL</name>
|
|
<description>the url of the postgress server to query</description>
|
|
</property>
|
|
<property>
|
|
<name>postgresUser</name>
|
|
<description>the username to access the postgres db</description>
|
|
</property>
|
|
<property>
|
|
<name>postgresPassword</name>
|
|
<description>the postgres password</description>
|
|
</property>
|
|
<property>
|
|
<name>sourcePath</name>
|
|
<description>the source path</description>
|
|
</property>
|
|
</parameters>
|
|
<start to="reset-outputpath"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
<action name="reset-outputpath">
|
|
<fs>
|
|
<delete path='${workingDir}/blacklist'/>
|
|
</fs>
|
|
<ok to="read_blacklist"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="read_blacklist">
|
|
<java>
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<main-class>eu.dnetlib.dhp.blacklist.ReadBlacklistFromDB</main-class>
|
|
<arg>--hdfsPath</arg><arg>${workingDir}/blacklist</arg>
|
|
<arg>--hdfsNameNode</arg><arg>${nameNode}</arg>
|
|
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
|
|
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
|
|
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
|
|
</java>
|
|
<ok to="prepare_merged_relation"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="prepare_merged_relation">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>PrepareMergedRelation</name>
|
|
<class>eu.dnetlib.dhp.blacklist.PrepareMergedRelationJob</class>
|
|
<jar>dhp-blacklist-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/mergesRelation</arg>
|
|
<arg>--hive_metastore_uris</arg><arg>${hive_metastore_uris}</arg>
|
|
</spark>
|
|
<ok to="apply_blacklist"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="apply_blacklist">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>ApplyBlacklist</name>
|
|
<class>eu.dnetlib.dhp.blacklist.SparkRemoveBlacklistedRelationJob</class>
|
|
<jar>dhp-blacklist-${projectVersion}.jar</jar>
|
|
<spark-opts>
|
|
--executor-cores=${sparkExecutorCores}
|
|
--executor-memory=${sparkExecutorMemory}
|
|
--driver-memory=${sparkDriverMemory}
|
|
--conf spark.extraListeners=${spark2ExtraListeners}
|
|
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
|
|
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
|
|
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
|
|
</spark-opts>
|
|
<arg>--sourcePath</arg><arg>${sourcePath}</arg>
|
|
<arg>--outputPath</arg><arg>${workingDir}/relation</arg>
|
|
<arg>--hdfsPath</arg><arg>${workingDir}/blacklist</arg>
|
|
<arg>--mergesPath</arg><arg>${workingDir}/mergesRelation</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
<end name="End"/>
|
|
</workflow-app> |