-werners-
Esteemed Contributor III

I have no idea tbh, but it seems like a bad idea. Databricks = Spark.

If you want to use Flink, I would run a Flink cluster.

If you are ok with spark, you do not need the pyflink library. Databricks has a Kakfa connector:

https://docs.databricks.com/structured-streaming/kafka.html

View solution in original post