dnet-hadoop/dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/raw_step2/oozie_app/workflow.xml

74 lines
3.1 KiB
XML

<workflow-app name="import regular entities as Graph (step 2)" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>migrationPathStep1</name>
<description>the base path to store hdfs file</description>
</property>
<property>
<name>migrationPathStep2</name>
<description>the temporary path to store entities before dispatching</description>
</property>
<property>
<name>postgresURL</name>
<description>the postgres URL to access to the database</description>
</property>
<property>
<name>postgresUser</name>
<description>the user postgres</description>
</property>
<property>
<name>postgresPassword</name>
<description>the password postgres</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>
</parameters>
<start to="ResetEntities"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResetEntities">
<fs>
<delete path='${migrationPathStep2}'/>
<mkdir path='${migrationPathStep2}'/>
</fs>
<ok to="GenerateEntities"/>
<error to="Kill"/>
</action>
<action name="GenerateEntities">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>GenerateEntities</name>
<class>eu.dnetlib.dhp.migration.step2.GenerateEntitiesApplication</class>
<jar>dhp-aggregation-${projectVersion}.jar</jar>
<spark-opts>--executor-memory ${sparkExecutorMemory} --executor-cores ${sparkExecutorCores} --driver-memory=${sparkDriverMemory} --conf spark.extraListeners="com.cloudera.spark.lineage.NavigatorAppListener" --conf spark.sql.queryExecutionListeners="com.cloudera.spark.lineage.NavigatorQueryListener" --conf spark.sql.warehouse.dir="/user/hive/warehouse"</spark-opts>
<arg>-mt</arg> <arg>yarn-cluster</arg>
<arg>-s</arg><arg>${migrationPathStep1}/db_records,${migrationPathStep1}/oaf_records,${migrationPathStep1}/odf_records</arg>
<arg>-t</arg><arg>${migrationPathStep2}/all_entities</arg>
<arg>-pgurl</arg><arg>${postgresURL}</arg>
<arg>-pguser</arg><arg>${postgresUser}</arg>
<arg>-pgpasswd</arg><arg>${postgresPassword}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>