Join Error

I get an error message when trying to join two different types of datasets. Admins have no problem, but I get this after I choose the fields to join on and click Apply. And both datasets are SPICE.

Hi @durl

When joining two datasets together from different sources it will default to SPICE and should allow you to do the join. Regardless of if they are datasets or other datasources.

I wonder if this is a permissions issue. Are you able to create a simple dataset using spice?

Ramon Lopez

Thanks for the reply.

Like you, I’m thinking it’s a permissions issue. An Admin can do the join, but I get the error. I’m having them check with other Users (non-admin) at the main office to see if they encounter the same problem.

I can create a simple database using SPICE, yes.

Hi drul,

You are encountering this error as mixing data from two datasets (from different sources) requires that the child dataset be in SPICE mode. There is a permission setting in parent datasets which controls whether SPICE copies are allowed or not. Please check / ask owners of parent datasets to check if those datasets are set to allow SPICE copies. To check this, on parent datasets, go to properties and check the state of ‘Allow SPICE copies’ under child datasets section (See screen shot below).


I’m marking this reply as solution here. Please feel free to let us know if you have more questions.

Arun Santhosh

Thank you for the reply. They looked and Allow SPICE copies is checked but the error about requiring SPICE to join is still there.

There’s a Redshift dataset (SPICE, with row-level security) I’m attempting to join to, but when attempt to use it for a new dataset, the Query Mode is locked at Direct.

Have you looked into this?

I’m going to pass this along and see if RLS was applied before SPICE.

Thank you.

1 Like