Use EMR serverless bundled iceberg JAR. (#2632)

Instead of downloading the JAR from Maven, the JAR in the EMR serverless root file system can be used. Signed-off-by: Adi Suresh <[email protected]> (cherry picked from commit e578a57) Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
opensearch-project · Apr 24, 2024 · 41e682f · 41e682f
1 parent 294566f
commit 41e682f
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 10 deletions.
diff --git a/spark/src/main/java/org/opensearch/sql/spark/asyncquery/model/SparkSubmitParameters.java b/spark/src/main/java/org/opensearch/sql/spark/asyncquery/model/SparkSubmitParameters.java
@@ -54,15 +54,10 @@ private Builder() {
       config.put(
           HADOOP_CATALOG_CREDENTIALS_PROVIDER_FACTORY_KEY,
           DEFAULT_GLUE_CATALOG_CREDENTIALS_PROVIDER_FACTORY_KEY);
+      config.put(SPARK_JARS_KEY, ICEBERG_SPARK_RUNTIME_PACKAGE);
       config.put(
           SPARK_JAR_PACKAGES_KEY,
-          SPARK_STANDALONE_PACKAGE
-              + ","
-              + SPARK_LAUNCHER_PACKAGE
-              + ","
-              + PPL_STANDALONE_PACKAGE
-              + ","
-              + ICEBERG_SPARK_RUNTIME_PACKAGE);
+          SPARK_STANDALONE_PACKAGE + "," + SPARK_LAUNCHER_PACKAGE + "," + PPL_STANDALONE_PACKAGE);
       config.put(SPARK_JAR_REPOSITORIES_KEY, AWS_SNAPSHOT_REPOSITORY);
       config.put(SPARK_DRIVER_ENV_JAVA_HOME_KEY, JAVA_HOME_LOCATION);
       config.put(SPARK_EXECUTOR_ENV_JAVA_HOME_KEY, JAVA_HOME_LOCATION);

diff --git a/spark/src/main/java/org/opensearch/sql/spark/data/constants/SparkConstants.java b/spark/src/main/java/org/opensearch/sql/spark/data/constants/SparkConstants.java
@@ -41,6 +41,7 @@ public class SparkConstants {
   public static final String HADOOP_CATALOG_CREDENTIALS_PROVIDER_FACTORY_KEY =
       "spark.hadoop.aws.catalog.credentials.provider.factory.class";
   public static final String HIVE_METASTORE_GLUE_ARN_KEY = "spark.hive.metastore.glue.role.arn";
+  public static final String SPARK_JARS_KEY = "spark.jars";
   public static final String SPARK_JAR_PACKAGES_KEY = "spark.jars.packages";
   public static final String SPARK_JAR_REPOSITORIES_KEY = "spark.jars.repositories";
   public static final String SPARK_DRIVER_ENV_JAVA_HOME_KEY =
@@ -101,7 +102,7 @@ public class SparkConstants {
   public static final String ICEBERG_SPARK_EXTENSION =
       "org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions";
   public static final String ICEBERG_SPARK_RUNTIME_PACKAGE =
-      "org.apache.iceberg:iceberg-spark-runtime-3.3_2.12:1.5.0";
+      "/usr/share/aws/iceberg/lib/iceberg-spark3-runtime.jar";
   public static final String SPARK_CATALOG_CATALOG_IMPL =
       "spark.sql.catalog.spark_catalog.catalog-impl";
   public static final String ICEBERG_GLUE_CATALOG = "org.apache.iceberg.aws.glue.GlueCatalog";

diff --git a/spark/src/test/java/org/opensearch/sql/spark/dispatcher/SparkQueryDispatcherTest.java b/spark/src/test/java/org/opensearch/sql/spark/dispatcher/SparkQueryDispatcherTest.java
@@ -949,8 +949,8 @@ private String constructExpectedSparkSubmitParameterString(
                + " spark.hadoop.fs.s3.customAWSCredentialsProvider=com.amazonaws.emr.AssumeRoleAWSCredentialsProvider"
                + "  --conf"
                + " spark.hadoop.aws.catalog.credentials.provider.factory.class=com.amazonaws.glue.catalog.metastore.STSAssumeRoleSessionCredentialsProviderFactory"
-               + "  --conf"
-               + " spark.jars.packages=org.opensearch:opensearch-spark-standalone_2.12:0.3.0-SNAPSHOT,org.opensearch:opensearch-spark-sql-application_2.12:0.3.0-SNAPSHOT,org.opensearch:opensearch-spark-ppl_2.12:0.3.0-SNAPSHOT,org.apache.iceberg:iceberg-spark-runtime-3.3_2.12:1.5.0"
+               + "  --conf spark.jars=/usr/share/aws/iceberg/lib/iceberg-spark3-runtime.jar  --conf"
+               + " spark.jars.packages=org.opensearch:opensearch-spark-standalone_2.12:0.3.0-SNAPSHOT,org.opensearch:opensearch-spark-sql-application_2.12:0.3.0-SNAPSHOT,org.opensearch:opensearch-spark-ppl_2.12:0.3.0-SNAPSHOT"
                + "  --conf"
                + " spark.jars.repositories=https://aws.oss.sonatype.org/content/repositories/snapshots"
                + "  --conf"