oozie workflow

This commit is contained in:
Michele Artini 2020-06-12 10:42:35 +02:00
parent c22cb5a3c6
commit 30ea1bda88
4 changed files with 146 additions and 2 deletions

View File

@ -51,7 +51,7 @@ public class GenerateEventsApplication {
IOUtils IOUtils
.toString( .toString(
GenerateEventsApplication.class GenerateEventsApplication.class
.getResourceAsStream("/eu/dnetlib/dhp/oa/graph/merge_claims_parameters.json"))); .getResourceAsStream("/eu/dnetlib/dhp/broker/oa/generate_broker_events.json")));
parser.parseArgument(args); parser.parseArgument(args);
final Boolean isSparkSessionManaged = Optional final Boolean isSparkSessionManaged = Optional
@ -149,7 +149,8 @@ public class GenerateEventsApplication {
return r4; return r4;
} }
private static <T, RT> Dataset<RT> relatedEntities(final Dataset<T> targets, final Dataset<Relation> rels, private static <T, RT> Dataset<RT> relatedEntities(final Dataset<T> targets,
final Dataset<Relation> rels,
final Class<RT> clazz) { final Class<RT> clazz) {
return rels return rels
.joinWith(targets, targets.col("id").equalTo(rels.col("target")), "inner") .joinWith(targets, targets.col("id").equalTo(rels.col("target")), "inner")

View File

@ -0,0 +1,18 @@
<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

@ -0,0 +1,99 @@
<workflow-app name="create broker events" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>graphInputPath</name>
<description>the path where the graph is stored</description>
</property>
<property>
<name>eventsOutputPath</name>
<description>the path where the the events will be stored</description>
</property>
<property>
<name>isLookupUrl</name>
<description>the address of the lookUp service</description>
</property>
<property>
<name>dedupConfProfId</name>
<description>the id of a valid Dedup Configuration Profile</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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="generate_events"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="generate_events">
<java>
<prepare>
<delete path="${eventsOutputPath}"/>
</prepare>
<main-class>eu.dnetlib.dhp.broker.oa.GenerateEventsApplication</main-class>
<arg>--graphPath</arg><arg>${graphInputPath}</arg>
<arg>--eventsPath</arg><arg>${eventsOutputPath}</arg>
<arg>--isLookupUrl</arg><arg>${isLookupUrl}</arg>
<arg>--dedupConfProfile</arg><arg>${dedupConfProfId}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,26 @@
[
{
"paramName": "g",
"paramLongName": "graphPath",
"paramDescription": "the path where there the graph is stored",
"paramRequired": true
},
{
"paramName": "o",
"paramLongName": "eventsPath",
"paramDescription": "the path where the generated events will be stored",
"paramRequired": true
},
{
"paramName": "lu",
"paramLongName": "isLookupUrl",
"paramDescription": "the address of the ISLookUpService",
"paramRequired": true
},
{
"paramName": "d",
"paramLongName": "dedupConfProfile",
"paramDescription": "the id of a valid Dedup Configuration Profile",
"paramRequired": true
}
]