45 lines
1.8 KiB
XML
45 lines
1.8 KiB
XML
<workflow-app name="Gen Enriched Orcid Works" xmlns="uri:oozie:workflow:0.5">
|
|
<parameters>
|
|
<property>
|
|
<name>workingPath</name>
|
|
<description>the working dir base path</description>
|
|
</property>
|
|
</parameters>
|
|
|
|
<start to="ResetWorkingPath"/>
|
|
|
|
<kill name="Kill">
|
|
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
|
|
</kill>
|
|
|
|
<action name="ResetWorkingPath">
|
|
<fs>
|
|
<delete path='${workingPath}/no_doi_enriched_works/output'/>
|
|
</fs>
|
|
<ok to="Gen_Enriched_Orcid_Works"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<action name="Gen_Enriched_Orcid_Works">
|
|
<spark xmlns="uri:oozie:spark-action:0.2">
|
|
<job-tracker>${jobTracker}</job-tracker>
|
|
<name-node>${nameNode}</name-node>
|
|
<master>yarn</master>
|
|
<mode>cluster</mode>
|
|
<name>Gen_Enriched_Orcid_Works</name>
|
|
<class>eu.dnetlib.doiboost.orcidnodoi.SparkGenEnrichedOrcidWorks</class>
|
|
<jar>dhp-doiboost-1.2.4-SNAPSHOT.jar</jar>
|
|
<spark-opts>--num-executors 10 --conf spark.yarn.jars="hdfs://hadoop-rm1.garr-pa1.d4science.org:8020/user/oozie/share/lib/lib_20180405103059/spark2" --executor-memory=${sparkExecutorMemory} --executor-cores=${sparkExecutorCores} --driver-memory=${sparkDriverMemory}
|
|
</spark-opts>
|
|
<arg>-w</arg><arg>${workingPath}/</arg>
|
|
<arg>-n</arg><arg>${nameNode}</arg>
|
|
<arg>-f</arg><arg>-</arg>
|
|
<arg>-ow</arg><arg>no_doi_works/</arg>
|
|
<arg>-oew</arg><arg>no_doi_enriched_works/output</arg>
|
|
</spark>
|
|
<ok to="End"/>
|
|
<error to="Kill"/>
|
|
</action>
|
|
|
|
<end name="End"/>
|
|
</workflow-app> |