cancel
Showing results forย 
Search instead forย 
Did you mean:ย 
Data Engineering
Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.
cancel
Showing results forย 
Search instead forย 
Did you mean:ย 

Databricks bug with show tblproperties - redacted - Azure databricks

SamDataWalk
New Contributor III

I am struggling to report what is a fairly fundamental bug. Can anyone help? Ideally someone from Databricks themselves. Or others who can confirm they can replicate it.

There is a bug where databricks seems to be hiding โ€œanyโ€ properties which have the characters โ€œuserโ€ in them.

This is not just in the display, but in the actual underlying result set.

 

To reproduce the bug.

 

create or replace view tmp_view

tblproperties (   notes='userid',   notes2='myuserid',   something='this will be fine')

as

select 1 as a;

 

 

Now in the next cell just run

 

show tblproperties tmp_view

 

and it doesnโ€™t just hide the data from the diusplay, but it changes the data inside the resultset. My application which connects to databricks and runs this command is actually getting the string *****(redacted)

 

 

1 ACCEPTED SOLUTION

Accepted Solutions

SamDataWalk
New Contributor III

I managed to get a response back from support at databricks.

Admittedly it is a bit nuclear, but there is a way of switching it off.

spark.conf.set("spark.databricks.behaviorChange.SC102534CommandRedactProperties.enabled", False)
So, I have managed to use this, to temporarily switch it off whilst this bit of my code is extracting the meta data.

View solution in original post

5 REPLIES 5

PatDBX
New Contributor II

I understand that yours is a View. For my case, it's a Table so I could use `desc detail <schema_name>.<table_name>` to get the table properties info that are not redacted in the `properties` column from the `desc detail` output.

SamDataWalk
New Contributor III

Yes, it is indeed a view which that command errors for.

thanks anyway, appreciate the suggestion 

SamDataWalk
New Contributor III

I managed to get a response back from support at databricks.

Admittedly it is a bit nuclear, but there is a way of switching it off.

spark.conf.set("spark.databricks.behaviorChange.SC102534CommandRedactProperties.enabled", False)
So, I have managed to use this, to temporarily switch it off whilst this bit of my code is extracting the meta data.

SamAdams
New Contributor III

@SamDataWalk thanks for posting the solution. I ran into the same problem while writing some logic that updates table constraints across many tables from a config. They were coming back as redacted - this allowed me to actually read them and update as needed.

SamAdams
New Contributor III

Like your example that redaction behavior seemed to pick up on the column name: a condition that included a column named "URL" was redacted, but one that included a "modifiedDateTime" was not

Connect with Databricks Users in Your Area

Join a Regional User Group to connect with local Databricks users. Events will be happening in your city, and you wonโ€™t want to miss the chance to attend and share knowledge.

If there isnโ€™t a group near you, start one and help create a community that brings people together.

Request a New Group