whitelist of topics

This commit is contained in:
Michele Artini 2020-12-14 11:03:55 +01:00
parent 467aa77279
commit 3c9cbd19f3
6 changed files with 158 additions and 2 deletions

View File

@ -59,6 +59,9 @@ public class GenerateEventsJob {
final Set<String> dsIdBlacklist = ClusterUtils.parseParamAsList(parser, "datasourceIdBlacklist");
log.info("datasourceIdBlacklist: {}", StringUtils.join(dsIdBlacklist, ","));
final Set<String> topicWhitelist = ClusterUtils.parseParamAsList(parser, "topicWhitelist");
log.info("topicWhitelist: {}", StringUtils.join(topicWhitelist, ","));
final SparkConf conf = new SparkConf();
runWithSparkSession(conf, isSparkSessionManaged, spark -> {
@ -75,7 +78,7 @@ public class GenerateEventsJob {
final Dataset<Event> dataset = groups
.map(
g -> EventFinder
.generateEvents(g, dsIdWhitelist, dsIdBlacklist, dsTypeWhitelist, accumulators),
.generateEvents(g, dsIdWhitelist, dsIdBlacklist, dsTypeWhitelist, topicWhitelist, accumulators),
Encoders
.bean(EventGroup.class))
.flatMap(g -> g.getData().iterator(), Encoders.bean(Event.class));

View File

@ -76,6 +76,7 @@ public class EventFinder {
final Set<String> dsIdWhitelist,
final Set<String> dsIdBlacklist,
final Set<String> dsTypeWhitelist,
final Set<String> topicWhitelist,
final Map<String, LongAccumulator> accumulators) {
final List<UpdateInfo<?>> list = new ArrayList<>();
@ -84,7 +85,13 @@ public class EventFinder {
for (final OaBrokerRelatedDatasource targetDs : target.getDatasources()) {
if (verifyTarget(targetDs, dsIdWhitelist, dsIdBlacklist, dsTypeWhitelist)) {
for (final UpdateMatcher<?> matcher : matchers) {
list.addAll(matcher.searchUpdatesForRecord(target, targetDs, results.getData(), accumulators));
for (final UpdateInfo<?> info : matcher
.searchUpdatesForRecord(target, targetDs, results.getData(), accumulators)) {
if (topicWhitelist == null || topicWhitelist.isEmpty()
|| topicWhitelist.contains(info.getTopic().getPath())) {
list.add(info);
}
}
}
}
}

View File

@ -24,6 +24,11 @@
<value>-</value>
<description>a black list (comma separeted, - for empty list) of datasource ids</description>
</property>
<property>
<name>topicWhitelist</name>
<value>*</value>
<description>a white list (comma separeted, * for all) of topics</description>
</property>
<property>
<name>esEventIndexName</name>
<description>the elasticsearch index name for events</description>
@ -447,6 +452,7 @@
<arg>--datasourceIdWhitelist</arg><arg>${datasourceIdWhitelist}</arg>
<arg>--datasourceTypeWhitelist</arg><arg>${datasourceTypeWhitelist}</arg>
<arg>--datasourceIdBlacklist</arg><arg>${datasourceIdBlacklist}</arg>
<arg>--topicWhitelist</arg><arg>${topicWhitelist}</arg>
</spark>
<ok to="index_event_subset"/>
<error to="Kill"/>

View File

@ -28,5 +28,11 @@
"paramLongName": "datasourceIdBlacklist",
"paramDescription": "a black list (comma separeted, - for empty list) of datasource ids",
"paramRequired": true
},
{
"paramName": "topicWhitelist",
"paramLongName": "topicWhitelist",
"paramDescription": "a white list (comma separeted, * for all) of topics",
"paramRequired": true
}
]

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 +1,116 @@
<workflow-app name="reindex_events" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>outputDir</name>
<description>the path where the the generated data will be stored</description>
</property>
<property>
<name>esEventIndexName</name>
<description>the elasticsearch index name for events</description>
</property>
<property>
<name>esIndexHost</name>
<description>the elasticsearch host</description>
</property>
<property>
<name>maxIndexedEventsForDsAndTopic</name>
<description>the max number of events for each couple (ds/topic)</description>
</property>
<property>
<name>brokerApiBaseUrl</name>
<description>the url of the broker service api</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>${oozieActionShareLibForSpark2}</value>
</property>
</configuration>
</global>
<start to="index_event_subset"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="index_event_subset">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>IndexEventSubsetOnESJob</name>
<class>eu.dnetlib.dhp.broker.oa.IndexEventSubsetJob</class>
<jar>dhp-broker-events-${projectVersion}.jar</jar>
<spark-opts>
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.dynamicAllocation.maxExecutors="8"
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.shuffle.partitions=3840
</spark-opts>
<arg>--outputDir</arg><arg>${outputDir}</arg>
<arg>--index</arg><arg>${esEventIndexName}</arg>
<arg>--esHost</arg><arg>${esIndexHost}</arg>
<arg>--maxEventsForTopic</arg><arg>${maxIndexedEventsForDsAndTopic}</arg>
<arg>--brokerApiBaseUrl</arg><arg>${brokerApiBaseUrl}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>