Closes #1426: Run IIS experiments by relying on spark 3.4 version

WIP. Fixing the changed results order in patent and software entity exporter integration tests. Introducing required fixes for various `iis-wf-export-actionmanager` exporters relying on spark3 to let their integration tests to succeed: * setting `spark.extraListeners` and `spark.sql.queryExecutionListeners` explicitly to empty values in order to avoid relying on incompatible, spark2 compliant, cloudera listeners * setting `spark.shuffle.useOldFetchProtocol=true` in order to address `2.4 to 3.0 migration guide` requirement regarding protocol for fetching shuffle blocks backward compatibility (and avoiding `IllegalArgumentException: Unexpected message type: <number>` kind of errors) The following modules were covered with similar workflow.xml related changes but their spark3 compatibility was not fully tested yet: * `iis-wf-affmatching` * `iis-wf-citationmatching-direct` * `iis-wf-citationmatching` * `iis-wf-documentsclassification` * `iis-wf-import` (`content_url/core_parquet`, `infospace`, `patent`) * `iis-wf-referenceextraction` (`community`, `concept`, `covid19`, `patent`, `project/funder_report`, `researchinitiative`, `softwareurl`) * `iis-wf-transformers` (`avro2json`)
openaire · May 8, 2024 · 337b8ef · 337b8ef
1 parent 2e40641
commit 337b8ef
Show file tree

Hide file tree

Showing 22 changed files with 41 additions and 59 deletions.
diff --git a/...affmatching/src/main/resources/eu/dnetlib/iis/wf/affmatching/dedup/oozie_app/workflow.xml b/...affmatching/src/main/resources/eu/dnetlib/iis/wf/affmatching/dedup/oozie_app/workflow.xml
@@ -84,7 +84,7 @@
 
     <action name="affmatching_dedup">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>document-organization deduplication</name>
             <class>eu.dnetlib.iis.wf.affmatching.AffMatchingDedupJob</class>

diff --git a/...-affmatching/src/main/resources/eu/dnetlib/iis/wf/affmatching/main/oozie_app/workflow.xml b/...-affmatching/src/main/resources/eu/dnetlib/iis/wf/affmatching/main/oozie_app/workflow.xml
@@ -122,7 +122,7 @@
     <action name="affiliation-matching">
         <spark xmlns="uri:oozie:spark-action:0.2">
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>Affiliation Matching</name>
             <class>eu.dnetlib.iis.wf.affmatching.AffMatchingJob</class>

diff --git a/...hing/src/main/resources/eu/dnetlib/iis/wf/affmatching/projectbased/oozie_app/workflow.xml b/...hing/src/main/resources/eu/dnetlib/iis/wf/affmatching/projectbased/oozie_app/workflow.xml
@@ -105,7 +105,7 @@
     <action name="proj-based-matching">
         <spark xmlns="uri:oozie:spark-action:0.2">
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>Project based Document-Organization Matching</name>
             <class>eu.dnetlib.iis.wf.affmatching.ProjectBasedMatchingJob</class>

diff --git a/...irect/src/main/resources/eu/dnetlib/iis/wf/citationmatching/direct/oozie_app/workflow.xml b/...irect/src/main/resources/eu/dnetlib/iis/wf/citationmatching/direct/oozie_app/workflow.xml
@@ -87,7 +87,7 @@
     <action name="citationmatchig-direct">
         <spark xmlns="uri:oozie:spark-action:0.2">
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>citationmatching_direct</name>
             <class>eu.dnetlib.iis.wf.citationmatching.direct.CitationMatchingDirectJob</class>

diff --git a/...tching/src/main/resources/eu/dnetlib/iis/wf/citationmatching/fuzzy/oozie_app/workflow.xml b/...tching/src/main/resources/eu/dnetlib/iis/wf/citationmatching/fuzzy/oozie_app/workflow.xml
@@ -103,7 +103,7 @@
 
     <action name="citation-matching-input-transformer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>citation-matching-input-transformer</name>
 
@@ -132,7 +132,7 @@
 
     <action name="citation-matching">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>citation-matching</name>
 
@@ -165,7 +165,7 @@
 
     <action name="citation-matching-output-transformer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>citation-matching-output-transformer</name>
 

diff --git a/...ation/src/main/resources/eu/dnetlib/iis/wf/documentsclassification/oozie_app/workflow.xml b/...ation/src/main/resources/eu/dnetlib/iis/wf/documentsclassification/oozie_app/workflow.xml
@@ -86,7 +86,7 @@
 	<action name="document_classification_job"> 
         <spark xmlns="uri:oozie:spark-action:0.2">
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
 
             <mode>cluster</mode>
 

diff --git a/...ain/resources/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/oozie_app/workflow.xml b/...ain/resources/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/oozie_app/workflow.xml
@@ -90,16 +90,6 @@
             <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>
@@ -133,7 +123,7 @@
 
     <action name="patent-exporter">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>patent spark exporter</name>
             <class>eu.dnetlib.iis.wf.export.actionmanager.entity.patent.PatentExporterJob</class>
@@ -142,8 +132,9 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
-                --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
+                --conf spark.extraListeners=
+                --conf spark.sql.queryExecutionListeners=
+                --conf spark.shuffle.useOldFetchProtocol=true
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
             </spark-opts>

diff --git a/...n/resources/eu/dnetlib/iis/wf/export/actionmanager/entity/software/oozie_app/workflow.xml b/...n/resources/eu/dnetlib/iis/wf/export/actionmanager/entity/software/oozie_app/workflow.xml
@@ -59,16 +59,6 @@
             <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>
@@ -103,7 +93,7 @@
     <action name="software-exporter">
         <spark xmlns="uri:oozie:spark-action:0.2">
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>software spark exporter</name>
             <class>eu.dnetlib.iis.wf.export.actionmanager.entity.software.SoftwareExporterJob</class>
@@ -113,8 +103,9 @@
                 --executor-memory=${sparkExecutorMemory}
                 --executor-cores=${sparkExecutorCores}
                 --driver-memory=${sparkDriverMemory}
-                --conf spark.extraListeners=${spark2ExtraListeners}
-                --conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
+                --conf spark.extraListeners=
+                --conf spark.sql.queryExecutionListeners=
+                --conf spark.shuffle.useOldFetchProtocol=true
                 --conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
                 --conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
             </spark-opts>

diff --git a/...urces/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/oozie_app/workflow.xml b/...urces/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/oozie_app/workflow.xml
@@ -101,12 +101,12 @@
             <arg>eu.dnetlib.iis.wf.export.actionmanager.sequencefile.TestingConsumer</arg>
             <arg>-Iseqfile=${workingDir}/output/document_patent/patent-actionset-id</arg>
             <arg>
-                -Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/document2_to_patent2.properties,
-                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent2_to_document2.properties,
+                -Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent2_to_document2.properties,
                 /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/document1_to_patent1.properties,
                 /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent1_to_document1.properties,
                 /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/document2_to_patent1.properties,
-                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent1_to_document2.properties
+                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent1_to_document2.properties,
+                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/document2_to_patent2.properties
             </arg>
         </java>
         <ok to="patent-entity-consumer"/>
@@ -119,8 +119,8 @@
             <arg>eu.dnetlib.iis.wf.export.actionmanager.sequencefile.TestingConsumer</arg>
             <arg>-Iseqfile=${workingDir}/output/entities_patent/patent-entity-actionset-id</arg>
             <arg>
-                -Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent1.expectations,
-                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent2.expectations
+                -Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent2.expectations,
+                /eu/dnetlib/iis/wf/export/actionmanager/entity/patent/default/output/patent1.expectations
             </arg>
         </java>
         <ok to="report-consumer"/>

diff --git a/...ces/eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/oozie_app/workflow.xml b/...ces/eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/oozie_app/workflow.xml
@@ -103,8 +103,8 @@
             <main-class>eu.dnetlib.iis.common.java.ProcessWrapper</main-class>
             <arg>eu.dnetlib.iis.wf.export.actionmanager.sequencefile.TestingConsumer</arg>
             <arg>-Iseqfile=${workingDir}/output/document_software_url/software-actionset-id</arg>
-            <arg>-Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/output/document_to_software.properties,
-                /eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/output/software_to_document.properties
+            <arg>-Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/output/software_to_document.properties,
+                /eu/dnetlib/iis/wf/export/actionmanager/entity/software/default/output/document_to_software.properties
             </arg>
         </java>
         <ok to="report-consumer" />

diff --git a/...es/eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/oozie_app/workflow.xml b/...es/eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/oozie_app/workflow.xml
@@ -103,8 +103,8 @@
             <main-class>eu.dnetlib.iis.common.java.ProcessWrapper</main-class>
             <arg>eu.dnetlib.iis.wf.export.actionmanager.sequencefile.TestingConsumer</arg>
             <arg>-Iseqfile=${workingDir}/output/document_software_url/software-actionset-id</arg>
-            <arg>-Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/output/document_to_software.properties,
-                /eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/output/software_to_document.properties
+            <arg>-Pexpectation_file_paths=/eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/output/software_to_document.properties,
+                /eu/dnetlib/iis/wf/export/actionmanager/entity/software/heritage/output/document_to_software.properties
             </arg>
         </java>
         <ok to="report-consumer" />

diff --git a/...main/resources/eu/dnetlib/iis/wf/importer/content_url/core_parquet/oozie_app/workflow.xml b/...main/resources/eu/dnetlib/iis/wf/importer/content_url/core_parquet/oozie_app/workflow.xml
@@ -82,7 +82,7 @@
 
     <action name="content-url-importer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>import-content-url_core</name>
             <class>eu.dnetlib.iis.wf.importer.content.HiveBasedDocumentContentUrlImporterJob</class>

diff --git a/...-wf-import/src/main/resources/eu/dnetlib/iis/wf/importer/infospace/oozie_app/workflow.xml b/...-wf-import/src/main/resources/eu/dnetlib/iis/wf/importer/infospace/oozie_app/workflow.xml
@@ -178,7 +178,7 @@
 
     <action name="infospace-importer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>import-infospace-hdp</name>
             <class>eu.dnetlib.iis.wf.importer.infospace.ImportInformationSpaceJob</class>

diff --git a/...iis-wf-import/src/main/resources/eu/dnetlib/iis/wf/importer/patent/oozie_app/workflow.xml b/...iis-wf-import/src/main/resources/eu/dnetlib/iis/wf/importer/patent/oozie_app/workflow.xml
@@ -77,7 +77,7 @@
 
     <action name="patent-importer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>import-patent</name>
             <class>eu.dnetlib.iis.wf.importer.patent.PatentReaderJob</class>

diff --git a/...ain/resources/eu/dnetlib/iis/wf/referenceextraction/community/main/oozie_app/workflow.xml b/...ain/resources/eu/dnetlib/iis/wf/referenceextraction/community/main/oozie_app/workflow.xml
@@ -62,7 +62,7 @@
                 </property>
             </configuration>
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>community-referenceextraction-input-transformer</name>
 

diff --git a/...u/dnetlib/iis/wf/referenceextraction/concept/root_conceptid_report/oozie_app/workflow.xml b/...u/dnetlib/iis/wf/referenceextraction/concept/root_conceptid_report/oozie_app/workflow.xml
@@ -67,7 +67,7 @@
                 </property>
             </configuration>
 
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>referenceextraction_root_conceptid_report</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.concept.RootConceptIdReportJob</class>

diff --git a/.../main/resources/eu/dnetlib/iis/wf/referenceextraction/covid19/main/oozie_app/workflow.xml b/.../main/resources/eu/dnetlib/iis/wf/referenceextraction/covid19/main/oozie_app/workflow.xml
@@ -93,7 +93,7 @@
                     <value>${oozieActionShareLibForSpark2}</value>
                 </property>
             </configuration>
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>covid19-referenceextraction-input-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.covid19.input.Covid19ReferenceExtractionInputTransformerJob</class>
@@ -185,7 +185,7 @@
                     <value>${oozieActionShareLibForSpark2}</value>
                 </property>
             </configuration>
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>covid19-referenceextraction-output-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.covid19.output.Covid19ToConceptIdTransformerJob</class>

diff --git a/...c/main/resources/eu/dnetlib/iis/wf/referenceextraction/patent/main/oozie_app/workflow.xml b/...c/main/resources/eu/dnetlib/iis/wf/referenceextraction/patent/main/oozie_app/workflow.xml
@@ -160,7 +160,7 @@
 
     <action name="input_transformer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>patent-referenceextraction-input-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.patent.input.PatentReferenceExtractionInputTransformerJob</class>
@@ -225,7 +225,7 @@
 
     <action name="metadata_retriever_input_transformer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>patent-metadata-retriever-input-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.patent.input.PatentMetadataRetrieverInputTransformerJob</class>
@@ -249,7 +249,7 @@
 
     <action name="metadata_retriever">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>patent-metadata-retriever</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.patent.PatentMetadataRetrieverJob</class>
@@ -292,7 +292,7 @@
 
     <action name="metadata_extractor">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>patent-metadata-extractor</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.patent.PatentMetadataExtractorJob</class>

diff --git a/...ources/eu/dnetlib/iis/wf/referenceextraction/project/funder_report/oozie_app/workflow.xml b/...ources/eu/dnetlib/iis/wf/referenceextraction/project/funder_report/oozie_app/workflow.xml
@@ -76,7 +76,7 @@
                     <value>${oozieActionShareLibForSpark2}</value>
                 </property>
             </configuration>
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>referenceextraction_project_funding_report</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.project.ProjectFunderReportJob</class>

diff --git a/...rces/eu/dnetlib/iis/wf/referenceextraction/researchinitiative/main/oozie_app/workflow.xml b/...rces/eu/dnetlib/iis/wf/referenceextraction/researchinitiative/main/oozie_app/workflow.xml
@@ -76,7 +76,7 @@
                     <value>${oozieActionShareLibForSpark2}</value>
                 </property>
             </configuration>
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>researchinitiative-input-metadata-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.researchinitiative.ResearchInitiativeReferenceExtractionInputTransformerJob</class>
@@ -107,7 +107,7 @@
                     <value>${oozieActionShareLibForSpark2}</value>
                 </property>
             </configuration>
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>researchinitiative-concept-transformer</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.researchinitiative.ResearchInitiativeMetadataTransformerJob</class>

diff --git a/...n/resources/eu/dnetlib/iis/wf/referenceextraction/softwareurl/main/oozie_app/workflow.xml b/...n/resources/eu/dnetlib/iis/wf/referenceextraction/softwareurl/main/oozie_app/workflow.xml
@@ -202,7 +202,7 @@
 
     <action name="cached-webcrawl">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>referenceextraction_softwareurl_webcrawl</name>
             <class>eu.dnetlib.iis.wf.referenceextraction.softwareurl.CachedWebCrawlerJob</class>

diff --git a/...ormers/src/main/resources/eu/dnetlib/iis/wf/transformers/avro2json/oozie_app/workflow.xml b/...ormers/src/main/resources/eu/dnetlib/iis/wf/transformers/avro2json/oozie_app/workflow.xml
@@ -79,7 +79,7 @@
 
     <action name="transformer">
         <spark xmlns="uri:oozie:spark-action:0.2">
-            <master>yarn-cluster</master>
+            <master>yarn</master>
             <mode>cluster</mode>
             <name>avro2json_transformer_job</name>
             <class>eu.dnetlib.iis.wf.transformers.avro2json.Avro2JsonTransformer</class>