1
0
Fork 0

followup changes reorganising the mdstore synchronisation mechanism

This commit is contained in:
Claudio Atzori 2023-03-01 10:16:20 +01:00
parent 49737f1087
commit 9c59dac859
14 changed files with 19 additions and 849 deletions

View File

@ -501,7 +501,6 @@ case object Crossref2Oaf {
queue += generateRelation(sourceId, targetId, ModelConstants.IS_PRODUCED_BY)
queue += generateRelation(targetId, sourceId, ModelConstants.PRODUCES)
case "10.13039/501100005416" => generateSimpleRelationFromAward(funder, "rcn_________", a => a)
case "10.13039/501100002848" => generateSimpleRelationFromAward(funder, "conicytf____", a => a)
case "10.13039/501100003448" => generateSimpleRelationFromAward(funder, "gsrt________", extractECAward)

View File

@ -214,16 +214,13 @@
<action name="ImportODF_claims">
<java>
<prepare>
<delete path="${contentPath}/odf_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${contentPath}/odf_claims</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>ODF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>claim</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/mdstore</arg>
<arg>--mongoBaseUrl</arg><arg>${mongoURL}</arg>
<arg>--mongoDb</arg><arg>${mongoDb}</arg>
<arg>--mdFormat</arg><arg>ODF</arg>
<arg>--mdLayout</arg><arg>store</arg>
<arg>--mdInterpretation</arg><arg>claim</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
</java>
<ok to="reuse_oaf_claims"/>
@ -240,16 +237,13 @@
<action name="ImportOAF_claims">
<java>
<prepare>
<delete path="${contentPath}/oaf_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${contentPath}/oaf_claims</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>OAF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>claim</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/mdstore</arg>
<arg>--mongoBaseUrl</arg><arg>${mongoURL}</arg>
<arg>--mongoDb</arg><arg>${mongoDb}</arg>
<arg>--mdFormat</arg><arg>OAF</arg>
<arg>--mdLayout</arg><arg>store</arg>
<arg>--mdInterpretation</arg><arg>claim</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
</java>
<ok to="wait_import"/>
@ -293,11 +287,8 @@
<action name="ImportODF">
<java>
<prepare>
<delete path="${contentPath}/odf_records"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/odf_records</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/mdstore</arg>
<arg>--mongoBaseUrl</arg><arg>${mongoURL}</arg>
<arg>--mongoDb</arg><arg>${mongoDb}</arg>
<arg>--mdFormat</arg><arg>ODF</arg>
@ -319,11 +310,8 @@
<action name="ImportOAF">
<java>
<prepare>
<delete path="${contentPath}/oaf_records"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/oaf_records</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/mdstore</arg>
<arg>--mongoBaseUrl</arg><arg>${mongoURL}</arg>
<arg>--mongoDb</arg><arg>${mongoDb}</arg>
<arg>--mdFormat</arg><arg>OAF</arg>
@ -337,11 +325,8 @@
<action name="ImportOAF_invisible">
<java>
<prepare>
<delete path="${contentPath}/oaf_records_invisible"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/oaf_records_invisible</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/mdstore</arg>
<arg>--mongoBaseUrl</arg><arg>${mongoURL}</arg>
<arg>--mongoDb</arg><arg>${mongoDb}</arg>
<arg>--mdFormat</arg><arg>OAF</arg>
@ -377,7 +362,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--hdfsPath</arg><arg>${contentPath}/odf_records_hdfs</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/odf_mdstore_hdfs</arg>
<arg>--mdstoreManagerUrl</arg><arg>${mdstoreManagerUrl}</arg>
<arg>--mdFormat</arg><arg>ODF</arg>
<arg>--mdLayout</arg><arg>store</arg>
@ -411,7 +396,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--hdfsPath</arg><arg>${contentPath}/oaf_records_hdfs</arg>
<arg>--hdfsPath</arg><arg>${contentPath}/oaf_mdstore_hdfs</arg>
<arg>--mdstoreManagerUrl</arg><arg>${mdstoreManagerUrl}</arg>
<arg>--mdFormat</arg><arg>OAF</arg>
<arg>--mdLayout</arg><arg>store</arg>
@ -544,7 +529,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_records,${contentPath}/odf_records,${contentPath}/oaf_records_hdfs,${contentPath}/odf_records_hdfs,${contentPath}/oaf_records_invisible</arg>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/mdstore/*/*</arg>
<arg>--invalidPath</arg><arg>${workingDir}/invalid_records</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
</spark>
@ -568,7 +553,7 @@
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_records,${contentPath}/odf_records,${contentPath}/oaf_records_hdfs,${contentPath}/odf_records_hdfs,${contentPath}/oaf_records_invisible</arg>
<arg>--sourcePaths</arg><arg>${contentPath}/db_openaire,${contentPath}/db_openorgs,${contentPath}/oaf_mdstore_hdfs,${contentPath}/odf_mdstore_hdfs,${contentPath}/mdstore/*/*</arg>
<arg>--targetPath</arg><arg>${workingDir}/entities</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--shouldHashId</arg><arg>${shouldHashId}</arg>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,162 +0,0 @@
<workflow-app name="import Claims as Graph" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>reuseContent</name>
<value>false</value>
<description>should import content from the aggregator or reuse a previous version</description>
</property>
<property>
<name>contentPath</name>
<description>path location to store (or reuse) content from the aggregator</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>dbSchema</name>
<value>beta</value>
<description>the database schema according to the D-Net infrastructure (beta or production)</description>
</property>
<property>
<name>mongoURL</name>
<description>mongoDB url, example: mongodb://[username:password@]host[:port]</description>
</property>
<property>
<name>mongoDb</name>
<description>mongo database</description>
</property>
<property>
<name>isLookupUrl</name>
<description>the address of the lookUp service</description>
</property>
<property>
<name>nsPrefixBlacklist</name>
<value></value>
<description>a blacklist of nsprefixes (comma separeted)</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>
<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>
<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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="ImportDB_claims"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ImportDB_claims">
<java>
<prepare>
<delete path="${contentPath}/db_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateDbEntitiesApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/db_claims</arg>
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--action</arg><arg>claims</arg>
<arg>--dbschema</arg><arg>${dbSchema}</arg>
<arg>--nsPrefixBlacklist</arg><arg>${nsPrefixBlacklist}</arg>
</java>
<ok to="ImportODF_claims"/>
<error to="Kill"/>
</action>
<action name="ImportODF_claims">
<java>
<prepare>
<delete path="${contentPath}/odf_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${contentPath}/odf_claims</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>ODF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>claim</arg>
</java>
<ok to="ImportOAF_claims"/>
<error to="Kill"/>
</action>
<action name="ImportOAF_claims">
<java>
<prepare>
<delete path="${contentPath}/oaf_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${contentPath}/oaf_claims</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>OAF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>claim</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,195 +0,0 @@
<workflow-app name="import DB entities" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>contentPath</name>
<description>path location to store (or reuse) content from the aggregator</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>dbSchema</name>
<value>beta</value>
<description>the database schema according to the D-Net infrastructure (beta or production)</description>
</property>
<property>
<name>isLookupUrl</name>
<description>the address of the lookUp service</description>
</property>
<property>
<name>nsPrefixBlacklist</name>
<value></value>
<description>a blacklist of nsprefixes (comma separeted)</description>
</property>
<property>
<name>reuseContent</name>
<value>false</value>
<description>reuse content in the aggregator database</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>
<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>
<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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="reuse_db"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<decision name="reuse_db">
<switch>
<case to="ImportDB">${wf:conf('reuseContent') eq false}</case>
<case to="GenerateEntities">${wf:conf('reuseContent') eq true}</case>
<default to="ImportDB"/>
</switch>
</decision>
<action name="ImportDB">
<java>
<prepare>
<delete path="${contentPath}/db_records"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateDbEntitiesApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/db_records</arg>
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--action</arg><arg>openaire</arg>
<arg>--dbschema</arg><arg>${dbSchema}</arg>
<arg>--nsPrefixBlacklist</arg><arg>${nsPrefixBlacklist}</arg>
</java>
<ok to="ImportDB_claims"/>
<error to="Kill"/>
</action>
<action name="ImportDB_claims">
<java>
<prepare>
<delete path="${contentPath}/db_claims"/>
</prepare>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateDbEntitiesApplication</main-class>
<arg>--hdfsPath</arg><arg>${contentPath}/db_claims</arg>
<arg>--postgresUrl</arg><arg>${postgresURL}</arg>
<arg>--postgresUser</arg><arg>${postgresUser}</arg>
<arg>--postgresPassword</arg><arg>${postgresPassword}</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--dbschema</arg><arg>${dbSchema}</arg>
<arg>--action</arg><arg>claims</arg>
<arg>--nsPrefixBlacklist</arg><arg>${nsPrefixBlacklist}</arg>
</java>
<ok to="GenerateEntities"/>
<error to="Kill"/>
</action>
<action name="GenerateEntities">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>GenerateEntities</name>
<class>eu.dnetlib.dhp.oa.graph.raw.GenerateEntitiesApplication</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory ${sparkExecutorMemory}
--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}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/db_records,${contentPath}/db_claims</arg>
<arg>--targetPath</arg><arg>${workingDir}/entities</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--shouldHashId</arg><arg>true</arg>
</spark>
<ok to="GenerateGraph"/>
<error to="Kill"/>
</action>
<action name="GenerateGraph">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>GenerateGraph</name>
<class>eu.dnetlib.dhp.oa.graph.raw.DispatchEntitiesApplication</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory ${sparkExecutorMemory}
--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}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=7680
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/entities</arg>
<arg>--graphRawPath</arg><arg>${workingDir}/graph_aggregator</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,157 +0,0 @@
<workflow-app name="Test Import of Hdfs Stores" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>graphOutputPath</name>
<description>the target path to store raw graph</description>
</property>
<property>
<name>contentPath</name>
<description>path location to store (or reuse) content from the aggregator</description>
</property>
<property>
<name>mdstoreManagerUrl</name>
<description>the address of the Mdstore Manager</description>
</property>
<property>
<name>isLookupUrl</name>
<description>the address of the lookUp service</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>
<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>
<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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="ImportODF_hdfs"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ImportODF_hdfs">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>ImportODF_hdfs</name>
<class>eu.dnetlib.dhp.oa.graph.raw.MigrateHdfsMdstoresApplication</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory ${sparkExecutorMemory}
--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}${spark2EventLogDir}
</spark-opts>
<arg>--hdfsPath</arg><arg>${contentPath}/odf_records_hdfs</arg>
<arg>--mdstoreManagerUrl</arg><arg>${mdstoreManagerUrl}</arg>
<arg>--mdFormat</arg><arg>ODF</arg>
<arg>--mdLayout</arg><arg>store</arg>
<arg>--mdInterpretation</arg><arg>cleaned</arg>
</spark>
<ok to="GenerateEntities"/>
<error to="Kill"/>
</action>
<action name="GenerateEntities">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>GenerateEntities</name>
<class>eu.dnetlib.dhp.oa.graph.raw.GenerateEntitiesApplication</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory ${sparkExecutorMemory}
--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}${spark2EventLogDir}
</spark-opts>
<arg>--sourcePaths</arg><arg>${contentPath}/odf_records_hdfs</arg>
<arg>--targetPath</arg><arg>${workingDir}/entities</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--shouldHashId</arg><arg>${shouldHashId}</arg>
</spark>
<ok to="GenerateGraph"/>
<error to="Kill"/>
</action>
<action name="GenerateGraph">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>GenerateGraph</name>
<class>eu.dnetlib.dhp.oa.graph.raw.DispatchEntitiesApplication</class>
<jar>dhp-graph-mapper-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory ${sparkExecutorMemory}
--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}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=7680
</spark-opts>
<arg>--sourcePath</arg><arg>${workingDir}/entities</arg>
<arg>--graphRawPath</arg><arg>${workingDir}/graph_raw</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,67 +0,0 @@
<workflow-app name="import regular entities as Graph (step 1)" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>migrationPathStep1</name>
<description>the base path to store hdfs file</description>
</property>
<property>
<name>mongoURL</name>
<description>mongoDB url, example: mongodb://[username:password@]host[:port]</description>
</property>
<property>
<name>mongoDb</name>
<description>mongo database</description>
</property>
</parameters>
<start to="ResetWorkingPath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResetWorkingPath">
<fs>
<delete path='${migrationPathStep1}'/>
<mkdir path='${migrationPathStep1}'/>
</fs>
<ok to="ImportODF"/>
<error to="Kill"/>
</action>
<action name="ImportODF">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${migrationPathStep1}</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>ODF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>cleaned</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
</java>
<ok to="ImportOAF"/>
<error to="Kill"/>
</action>
<action name="ImportOAF">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.dhp.oa.graph.raw.MigrateMongoMdstoresApplication</main-class>
<arg>-p</arg><arg>${migrationPathStep1}</arg>
<arg>-mongourl</arg><arg>${mongoURL}</arg>
<arg>-mongodb</arg><arg>${mongoDb}</arg>
<arg>-f</arg><arg>OAF</arg>
<arg>-l</arg><arg>store</arg>
<arg>-i</arg><arg>cleaned</arg>
<arg>--nameNode</arg><arg>${nameNode}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,65 +0,0 @@
<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>isLookupUrl</name>
<description>the address of the lookUp service</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>--islookup</arg><arg>${isLookupUrl}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -1,18 +0,0 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

View File

@ -1,60 +0,0 @@
<workflow-app name="import regular entities as Graph (step 3)" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>migrationPathStep2</name>
<description>the temporary path to store entities before dispatching</description>
</property>
<property>
<name>migrationPathStep3</name>
<description>the graph Raw base path</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="ResetGraph"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResetGraph">
<fs>
<delete path='${migrationPathStep3}'/>
<mkdir path='${migrationPathStep3}'/>
</fs>
<ok to="GenerateGraph"/>
<error to="Kill"/>
</action>
<action name="GenerateGraph">
<spark xmlns="uri:oozie:spark-action:0.2">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>GenerateGraph</name>
<class>eu.dnetlib.dhp.migration.step3.DispatchEntitiesApplication</class>
<jar>dhp-aggregation-${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>${migrationPathStep2}/all_entities</arg>
<arg>-g</arg><arg>${migrationPathStep3}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>