From 3101ab8eaa44f3a20901d3da1274a4dbe64464fe Mon Sep 17 00:00:00 2001 From: Jonas Thelemann Date: Wed, 21 Aug 2024 17:18:05 +0200 Subject: [PATCH] docs(spark-setup): improve namespace caveat description A more detailed description would've saved me quite a bit of time at dbt setup. --- website/docs/docs/core/connect-data-platform/spark-setup.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/website/docs/docs/core/connect-data-platform/spark-setup.md b/website/docs/docs/core/connect-data-platform/spark-setup.md index 3b1429c246b..9f24f7a5a7c 100644 --- a/website/docs/docs/core/connect-data-platform/spark-setup.md +++ b/website/docs/docs/core/connect-data-platform/spark-setup.md @@ -208,6 +208,8 @@ Spark can be customized using [Application Properties](https://spark.apache.org/ ## Caveats +When facing difficulties you can run `poetry run dbt debug --log-level=debug`. The logs are persisted at `logs/dbt.log`. + ### Usage with EMR To connect to Apache Spark running on an Amazon EMR cluster, you will need to run `sudo /usr/lib/spark/sbin/start-thriftserver.sh` on the master node of the cluster to start the Thrift server (see [the docs](https://aws.amazon.com/premiumsupport/knowledge-center/jdbc-connection-emr/) for more information). You will also need to connect to port 10001, which will connect to the Spark backend Thrift server; port 10000 will instead connect to a Hive backend, which will not work correctly with dbt. @@ -223,6 +225,6 @@ Delta-only features: ### Default namespace with Thrift connection method -If your Spark cluster doesn't have a default namespace, metadata queries that run before any dbt workflow will fail, causing the entire workflow to fail, even if your configurations are correct. The metadata queries fail there's no default namespace in which to run it. +A namespace named `default` is required to exist in Spark when connecting via Thrift for dbt to run metadata queries in. You can use Spark's `pyspark` and run `spark.sql("SHOW NAMESPACES").show()` to see the available namespaces and create the required namespace by running `spark.sql("CREATE NAMESPACE default").show()`. -To debug, review the debug-level logs to confirm the query dbt is running when it encounters the error: `dbt run --debug` or `logs/dbt.log`. +If there's a network connection issue instead, your logs will contain `Could not connect to any of [('127.0.0.1', 10000)]` (or similar).