dbt error: Data too long for column at row 1

[611124]

Hi there!

We are experiencing a Databricks error we don’t recognise when we are running one of our event-based dbt models in dbt core (version 1.6.18). The dbt model uses the ‘insert_by_period’ materialisation that is still experimental for version 1.6.18.

The error message states:

 Error during request to server: (conn=4757514) Data too long for column 'session_data' at row 1

We could narrow down the timeframe of the events in which this error occurs but not the error itself, since there is no column called ‘session_data’ . Has anyone encountered the same error before and knows what the issue might be?

We hope that we won’t get the same error message again once we have upgraded to dbt version 1.9.1, since the ‘insert_by_period’ period won’t be experimental anymore but it would be good to get to the bottom of this until then.

Thanks!

[NandiniN]

This KB https://kb.databricks.com/metastore/data-too-long-for-column documents the cause and solution.

Can you please check the table definition if session_data is an implicit column added.

[611124]

Hi there! Thank you for your reply and the link, I'll take a look. 

Unfortunately, session_data isn't an implicit column either.

[NandiniN]

Can you please try if you are seeing this error with the latest dbt version?

[611124]

We are yet to upgrade dbt core to the latest version but will check again once we have done so.