andreas.czerniak
/
BrBETA_dnet-hadoop
forked from D-Net/dnet-hadoop
1
1
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

adjusted parameters

This commit is contained in:
Claudio Atzori 2020-05-04 19:18:59 +02:00
parent f1b7e14036
commit a2fc37df5f
1 changed files with 68 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

85
dhp-workflows/dhp-graph-mapper/src/main/resources/eu/dnetlib/dhp/oa/graph/raw_db/oozie_app/workflow.xml
View File

@ -1,8 +1,8 @@
<workflow-app name="import db entities (step 1)" xmlns="uri:oozie:workflow:0.5"> <workflow-app name="import DB entities" xmlns="uri:oozie:workflow:0.5">
<parameters> <parameters>
<property> <property>
<name>migrationPathStep1</name> <name>contentPath</name>
<description>the base path to store hdfs file</description> <description>path location to store (or reuse) content from the aggregator</description>
</property> </property>
<property> <property>
<name>postgresURL</name> <name>postgresURL</name>
@ -16,6 +16,7 @@
<name>postgresPassword</name> <name>postgresPassword</name>
<description>the password postgres</description> <description>the password postgres</description>
</property> </property>
<property> <property>
<name>sparkDriverMemory</name> <name>sparkDriverMemory</name>
<description>memory for driver process</description> <description>memory for driver process</description>
@ -28,31 +29,81 @@
<name>sparkExecutorCores</name> <name>sparkExecutorCores</name>
<description>number of cores used by single executor</description> <description>number of cores used by single executor</description>
</property> </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> </parameters>
<start to="ResetWorkingPath"/> <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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="ImportDB"/>
<kill name="Kill"> <kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message> <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill> </kill>
<action name="ResetWorkingPath"> <action name="ImportDB">
<fs> <java>
<delete path='${migrationPathStep1}/db_records'/> <prepare>
</fs> <delete path="${contentPath}/db_records"/>
<ok to="ImportDB"/> </prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateDbEntitiesApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/db_records</arg>
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
</java>
<ok to="ImportDB_claims"/>
<error to="Kill"/> <error to="Kill"/>
</action> </action>
<action name="ImportDB"> <action name="ImportDB_claims">
<java> <java>
<job-tracker>${jobTracker}</job-tracker> <prepare>
<name-node>${nameNode}</name-node> <delete path="${contentPath}/db_claims"/>
<main-class>eu.dnetlib.dhp.migration.step1.MigrateDbEntitiesApplication</main-class> </prepare>
<arg>-p</arg><arg>${migrationPathStep1}/db_records</arg> <main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateDbEntitiesApplication</main-class>
<arg>-pgurl</arg><arg>${postgresURL}</arg> <arg>--hdfsPath</arg><arg>${contentPath}/db_claims</arg>
<arg>-pguser</arg><arg>${postgresUser}</arg> <arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>-pgpasswd</arg><arg>${postgresPassword}</arg> <arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
<arg>--action</arg><arg>claims</arg>
</java> </java>
<ok to="End"/> <ok to="End"/>
<error to="Kill"/> <error to="Kill"/>