search cancel

DE_IDENT_IND column doesn't HIDE after masking

book

Article ID: 255874

calendar_today

Updated On:

Products

CA Test Data Manager (Data Finder / Grid Tools)

Issue/Introduction

Per the Masking Options, we set the HIDERESTARTCOLUMN option to Y for all our databases. However, when checking the Oracle tables, the DE_IDENT_IND column isn't getting set to be INVISIBLE. 

 



 

Environment

Release : 4.10

Cause

The HIDERESTARTCOLUMN option only pertain to Oracle Tables. This option does not work for any other Data source

 

The confusion around the HIDERESTARTCOLUMN is partially due to the option name and a lack of documentation around how this option works. Engineering  explained that the FDM Masking Engine is not setting the column as 'invisible' in the Oracle database. Instead, it is setting the column as 'unused'.
 
Below is an example of the SQL Statement that is executed:
 
alter table <table_name> set unused column <restart_column>
 
So in fact the table is marked as 'unused', which makes the column inaccessible in Oracle. The difference between the drop column and 'hidden/unused' is the column data is not removed from the database, as it would be if the table is dropped. The benefit of this is increased performance on huge tables. It's much faster to mark the column 'unused' than to drop the column.

Resolution

 Engineering agreed to the enhancement and said that this may require a new option/functionality to be developed and tested.

One caveat with the new INVISIBLE option is the DROPRESET option must be set to No, in order for the INVISIBLE option to work, otherwise, the column will be dropped by default.

Starting in FastDataMasker-4.10.76.0, the behavior of the HIDERESTARTCOLUMN option has been changed to set the column as 'invisible' instead of 'unused'.