Activate spark extensions on SQL Endpoints

[juanc]

It would be possible to activate a custom extensions like Sedona (https://sedona.apache.org/download/databricks/ ) in SQL Endopoints?

Example error:

java.lang.ClassNotFoundException: org.apache.spark.sql.sedona_sql.UDT.GeometryUDT at org.apache.spark.sql.hive.thriftserver.SparkExecuteStatementOperation.org$apache$spark$sql$hive$thrift'. (35) (SQLExecDirectW)

[Kaniz_Fatma]

Hi @JUAN MENDEZ​ , You just need to install the Sedona jars and Sedona Python on Databricks using Databricks' default web UI. Then everything will work.

Please have a look at the documentation.

[Brendon_Daugher]

@Kaniz Fatma​ does this work for SQL Endpoints as well as clusters?

I am able to install this extension using the Sedona jars with no problems on my compute cluster, but it seems to me that the UI doesn't have an option for me to do this for a SQL Warehouse. I am using azure, so I'm not sure if that makes it different.

For example, when I click into my compute cluster, I see the tab shown at the bottom, while on my "starter endpoint" I don't see an analagous option.

Thanks in advance!

[Arh]

Having the same issue. Working with sedona in a cluster is no problem. However, when using an SQL endpoint the following error occurs:

org.apache.spark.sql.sedona_sql.UDT.GeometryUDT

Im working in Azure Databricks (latest version)

Wondering if anyone has a solutution or workaround.

[marcin-sg]

Same here.

[wallfacer]

@Kaniz_Fatma any updates here?

[ergü]

same hier on Azure Databricks.

[naveenanto]

@Kaniz_Fatma What is the right way to add custom spark extension to sql warehouse clusters?