Minor updates and bug fixes for RDS source (opensearch-project#4887)

* Minor updates and bug fixes to prepare for performance testing Signed-off-by: Hai Yan <oeyh@amazon.com> * Address comments Signed-off-by: Hai Yan <oeyh@amazon.com> --------- Signed-off-by: Hai Yan <oeyh@amazon.com>
dlvenable · Aug 29, 2024 · 8eef2f6 · 8eef2f6
1 parent b5358b5
commit 8eef2f6
Show file tree

Hide file tree

Showing 7 changed files with 42 additions and 8 deletions.
diff --git a/...ns/rds-source/src/main/java/org/opensearch/dataprepper/plugins/source/rds/RdsService.java b/...ns/rds-source/src/main/java/org/opensearch/dataprepper/plugins/source/rds/RdsService.java
@@ -105,7 +105,7 @@ public void start(Buffer<Record<Event>> buffer) {
 
         if (sourceConfig.isStreamEnabled()) {
             BinaryLogClient binaryLogClient = new BinlogClientFactory(sourceConfig, rdsClient, dbMetadata).create();
-            if (sourceConfig.getTlsConfig() == null || !sourceConfig.getTlsConfig().isInsecure()) {
+            if (sourceConfig.isTlsEnabled()) {
                 binaryLogClient.setSSLMode(SSLMode.REQUIRED);
             } else {
                 binaryLogClient.setSSLMode(SSLMode.DISABLED);
@@ -146,7 +146,7 @@ private SchemaManager getSchemaManager(final RdsSourceConfig sourceConfig, final
                 dbMetadata.getPort(),
                 sourceConfig.getAuthenticationConfig().getUsername(),
                 sourceConfig.getAuthenticationConfig().getPassword(),
-                !sourceConfig.getTlsConfig().isInsecure());
+                sourceConfig.isTlsEnabled());
         return new SchemaManager(connectionManager);
     }
 }
diff --git a/...s-source/src/main/java/org/opensearch/dataprepper/plugins/source/rds/RdsSourceConfig.java b/...s-source/src/main/java/org/opensearch/dataprepper/plugins/source/rds/RdsSourceConfig.java
@@ -85,6 +85,9 @@ public class RdsSourceConfig {
     @JsonProperty("tls")
     private TlsConfig tlsConfig;
 
+    @JsonProperty("disable_s3_read_for_leader")
+    private boolean disableS3ReadForLeader = false;
+
     public String getDbIdentifier() {
         return dbIdentifier;
     }
@@ -153,6 +156,14 @@ public TlsConfig getTlsConfig() {
         return tlsConfig;
     }
 
+    public boolean isTlsEnabled() {
+        return tlsConfig == null || !tlsConfig.isInsecure();
+    }
+
+    public boolean isDisableS3ReadForLeader() {
+        return disableS3ReadForLeader;
+    }
+
     public AuthenticationConfig getAuthenticationConfig() {
         return this.authenticationConfig;
     }

diff --git a/...n/java/org/opensearch/dataprepper/plugins/source/rds/converter/StreamRecordConverter.java b/...n/java/org/opensearch/dataprepper/plugins/source/rds/converter/StreamRecordConverter.java
@@ -16,6 +16,7 @@
 import java.nio.ByteBuffer;
 import java.security.MessageDigest;
 import java.security.NoSuchAlgorithmException;
+import java.time.Instant;
 import java.util.List;
 import java.util.Map;
 import java.util.stream.Collectors;
@@ -63,6 +64,11 @@ public Event convert(final Map<String, Object> rowData,
 
         EventMetadata eventMetadata = event.getMetadata();
 
+        // Only set external origination time for stream events, not export
+        final Instant externalOriginationTime = Instant.ofEpochMilli(eventCreateTimeEpochMillis);
+        event.getEventHandle().setExternalOriginationTime(externalOriginationTime);
+        eventMetadata.setExternalOriginationTime(externalOriginationTime);
+
         eventMetadata.setAttribute(EVENT_DATABASE_NAME_METADATA_ATTRIBUTE, databaseName);
         eventMetadata.setAttribute(EVENT_TABLE_NAME_METADATA_ATTRIBUTE, tableName);
         eventMetadata.setAttribute(BULK_ACTION_METADATA_ATTRIBUTE, bulkAction.toString());

diff --git a/...ce/src/main/java/org/opensearch/dataprepper/plugins/source/rds/model/ExportObjectKey.java b/...ce/src/main/java/org/opensearch/dataprepper/plugins/source/rds/model/ExportObjectKey.java
@@ -36,7 +36,9 @@ public static ExportObjectKey fromString(final String objectKeyString) {
         final String prefix = parts[0];
         final String exportTaskId = parts[1];
         final String databaseName = parts[2];
-        final String tableName = parts[3];
+        // fullTableName is in the format of "databaseName.tableName"
+        final String fullTableName = parts[3];
+        final String tableName = fullTableName.split("\\.")[1];
         final String numberedFolder = parts[4];
         final String fileName = parts[5];
         return new ExportObjectKey(prefix, exportTaskId, databaseName, tableName, numberedFolder, fileName);

diff --git a/...e/src/main/java/org/opensearch/dataprepper/plugins/source/rds/stream/StreamScheduler.java b/...e/src/main/java/org/opensearch/dataprepper/plugins/source/rds/stream/StreamScheduler.java
@@ -22,6 +22,8 @@
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Executors;
 
+import static org.opensearch.dataprepper.model.source.s3.S3ScanEnvironmentVariables.STOP_S3_SCAN_PROCESSING_PROPERTY;
+
 
 public class StreamScheduler implements Runnable {
 
@@ -57,18 +59,24 @@ public StreamScheduler(final EnhancedSourceCoordinator sourceCoordinator,
     @Override
     public void run() {
         LOG.debug("Start running Stream Scheduler");
+        StreamPartition streamPartition = null;
         while (!shutdownRequested && !Thread.currentThread().isInterrupted()) {
             try {
                 final Optional<EnhancedSourcePartition> sourcePartition = sourceCoordinator.acquireAvailablePartition(StreamPartition.PARTITION_TYPE);
                 if (sourcePartition.isPresent()) {
                     LOG.info("Acquired partition to read from stream");
 
-                    final StreamPartition streamPartition = (StreamPartition) sourcePartition.get();
+                    if (sourceConfig.isDisableS3ReadForLeader()) {
+                        // Primary node that acquires the stream partition will not perform work on the S3 buffer
+                        System.setProperty(STOP_S3_SCAN_PROCESSING_PROPERTY, "true");
+                    }
+
+                    streamPartition = (StreamPartition) sourcePartition.get();
                     final StreamCheckpointer streamCheckpointer = new StreamCheckpointer(sourceCoordinator, streamPartition, pluginMetrics);
                     binaryLogClient.registerEventListener(new BinlogEventListener(
                             buffer, sourceConfig, pluginMetrics, binaryLogClient, streamCheckpointer, acknowledgementSetManager));
                     final StreamWorker streamWorker = StreamWorker.create(sourceCoordinator, binaryLogClient, pluginMetrics);
-                    executorService.submit(() -> streamWorker.processStream(streamPartition));
+                    executorService.submit(() -> streamWorker.processStream((StreamPartition) sourcePartition.get()));
                 }
 
                 try {
@@ -81,6 +89,13 @@ public void run() {
 
             } catch (Exception e) {
                 LOG.error("Received an exception during stream processing, backing off and retrying", e);
+                if (streamPartition != null) {
+                    if (sourceConfig.isDisableS3ReadForLeader()) {
+                        System.clearProperty(STOP_S3_SCAN_PROCESSING_PROPERTY);
+                    }
+                    sourceCoordinator.giveUpPartition(streamPartition);
+                }
+
                 try {
                     Thread.sleep(DEFAULT_TAKE_LEASE_INTERVAL_MILLIS);
                 } catch (final InterruptedException ex) {

diff --git a/...c/test/java/org/opensearch/dataprepper/plugins/source/rds/export/ExportSchedulerTest.java b/...c/test/java/org/opensearch/dataprepper/plugins/source/rds/export/ExportSchedulerTest.java
@@ -124,7 +124,7 @@ void test_given_export_partition_and_export_task_id_then_complete_export() throw
         String tableName = UUID.randomUUID().toString();
         // objectKey needs to have this structure: "{prefix}/{export task ID}/{database name}/{table name}/{numbered folder}/{file name}"
         S3Object s3Object = S3Object.builder()
-                .key("prefix/" + exportTaskId + "/my_db/" + tableName + "/1/file1" + PARQUET_SUFFIX)
+                .key("prefix/" + exportTaskId + "/my_db/my_db." + tableName + "/1/file1" + PARQUET_SUFFIX)
                 .build();
         when(listObjectsV2Response.contents()).thenReturn(List.of(s3Object));
         when(listObjectsV2Response.isTruncated()).thenReturn(false);
@@ -185,7 +185,7 @@ void test_given_export_partition_without_export_task_id_then_start_and_complete_
         String tableName = UUID.randomUUID().toString();
         // objectKey needs to have this structure: "{prefix}/{export task ID}/{database name}/{table name}/{numbered folder}/{file name}"
         S3Object s3Object = S3Object.builder()
-                .key("prefix/" + exportTaskId + "/my_db/" + tableName + "/1/file1" + PARQUET_SUFFIX)
+                .key("prefix/" + exportTaskId + "/my_db/my_db." + tableName + "/1/file1" + PARQUET_SUFFIX)
                 .build();
         when(listObjectsV2Response.contents()).thenReturn(List.of(s3Object));
         when(listObjectsV2Response.isTruncated()).thenReturn(false);

diff --git a/...rc/test/java/org/opensearch/dataprepper/plugins/source/rds/model/ExportObjectKeyTest.java b/...rc/test/java/org/opensearch/dataprepper/plugins/source/rds/model/ExportObjectKeyTest.java
@@ -16,7 +16,7 @@ class ExportObjectKeyTest {
 
     @Test
     void test_fromString_with_valid_input_string() {
-        final String objectKeyString = "prefix/export-task-id/db-name/table-name/1/file-name.parquet";
+        final String objectKeyString = "prefix/export-task-id/db-name/db-name.table-name/1/file-name.parquet";
         final ExportObjectKey exportObjectKey = ExportObjectKey.fromString(objectKeyString);
 
         assertThat(exportObjectKey.getPrefix(), equalTo("prefix"));