Skip to main content

​
Overview

For Spark, the OpenLineage integration is published to Maven. A comprehensive guide on how to configure and use the OpenLineage integration with Apache Spark can be found at OpenLineage documentation.

​
Getting Started

In this tutorial, you’ll be running a simple self-contained Spark application with Iceberg locally via Docker Compose. We’ll use an existing example iceberg table demo.nyc.taxis used in the Iceberg quickstart guide for Spark. Our team is based in SF, so naturally, we’ll be using demo.sf.waymo.

​
Clone

Clone our tutorial repository and navigate to the spark directory: 
git clone https://github.com/OleanderHQ/tutorials.git

​
Requirements

​
Oleander API Key

Go to https://oleander.dev, then copy your API key. In the step below, replace [OLEANDER-API-KEY] with your API key before running the Spark job.

​
Running the Spark Job

Next, start up the docker containers with: 
docker-compose up
Then, run: 
docker exec -it spark-iceberg /opt/spark/bin/spark-submit \
  --conf spark.jars.packages=io.openlineage:openlineage-spark_2.12:1.26.0 \
  --conf spark.extraListeners=io.openlineage.spark.agent.OpenLineageSparkListener \
  --conf spark.openlineage.transport.type=http \
  --conf spark.openlineage.transport.url=https://oleander.dev \
  --conf spark.openlineage.transport.auth.type=api_key \
  --conf spark.openlineage.transport.auth.apiKey=[OLEANDER-API-KEY] \
  --conf spark.openlineage.namespace=spark-with-iceberg \
  --conf spark.sql.extensions=org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions \
  --conf spark.sql.catalog.spark_catalog=org.apache.iceberg.spark.SparkSessionCatalog \
  --conf spark.sql.catalog.spark_catalog.type=hive \
  --conf spark.sql.catalog.local=org.apache.iceberg.spark.SparkCatalog \
  --conf spark.sql.catalog.local.type=hadoop \
  --conf spark.sql.catalog.local.warehouse=$PWD/warehouse \
  --conf spark.sql.defaultCatalog=local \
  spark-with-iceberg.py
Finally, go to https://oleander.dev to view the OpenLineage events emitted by your Spark job.
⌘I