Test Data Manager

  • 1.  Help with scramble option for masking in GT subset

    Posted Jun 01, 2018 12:46 AM

    We are trying to perform inflight masking of oracle database tables using scramble options in GT subset.

    We have already installed GTREP and SCRAMBLE database as well.

    When try to generate a script using transformation map and scramble option ,   we are getting an error message saying it is not able to fetch the data from SCRAMBLE database.

     

    Please note for repository we have used SQL



  • 2.  Re: Help with scramble option for masking in GT subset

    Broadcom Employee
    Posted Jun 01, 2018 10:24 AM

    Hi Velentina,

     

    This sounds like a matter of using the complete host DNS name in the connection string.  I would start there.  Then I would review this section of the documentation Generate Masking Scripts for SQL Server and make sure that you have done all items in there.

     

    If this does not clear up the issue, please feel free to open a support case.  This will allow for a greater freedom of information exchange that may not be appropriate for a community exchange.


    Cheers!
    Les.



  • 3.  Re: Help with scramble option for masking in GT subset

    Posted Jun 03, 2018 09:20 PM

    Hi Les,

    Thanks for the information.

    We find that the scramble masking options for Oracle database are different from that of SQL database.

    We have used "Build windows datapump masked export".

    Both the gtrep and scramble database are present in the same server, however we have connected the datamaker using gtrep and giving the default database as gtrep, after which we connect our source and target tables . For some reason , when we generate the script using GT Subset , it is trying to fetch the data from scramble database . This is where we failed.

    Is there some documentation for inflight masking of Oracle database and not SQL database.

     

     

    Regards

    Velentina



  • 4.  Re: Help with scramble option for masking in GT subset
    Best Answer

    Broadcom Employee
    Posted Jun 04, 2018 03:59 AM

    Velentina when you are using subset and masking for an Oracle database, the scramble database has to reside on the same Oracle db server as the schema(database) you are trying to subset and mask.  If you are using the script that exports/mask and you specify a transformation map that requires a scramble database (eg using seedtables etc) then a scramble db will be required



  • 5.  Re: Help with scramble option for masking in GT subset

    Posted Jun 06, 2018 08:46 PM

    Hi Billy,

    Thanks for the information.

    So do we need to install the scramble database on Both the source database as well as target database with the schema for scramble database same as that of our source and tarrget database.

    Thanks

    Velentina



  • 6.  Re: Help with scramble option for masking in GT subset

    Broadcom Employee
    Posted Jun 07, 2018 04:00 AM

    You just need it on the source.  We mask as we are dumping out the data.



  • 7.  Re: Help with scramble option for masking in GT subset

    Posted Jun 08, 2018 02:43 AM

    Hi Billy,

     

    We are facing the same issue. However we have additional question,when you mention; " the scramble database has to reside on the same Oracle db server as the schema(database) you are trying to subset and mask" Where should be installed scrambled database? understanding that we have 2 oracle instances over same database engine and we must mask 5 tables from schema A which is located in oracle instance X and 10 tables from schema B which is located in oracle instance Y. 

     

    Thanks,

     

    GA



  • 8.  Re: Help with scramble option for masking in GT subset

    Broadcom Employee
    Posted Jun 10, 2018 02:09 PM

    It is my understanding that an Oracle instance can only have 1 database mounted/connected too at any one time.  Therefore I am assuming that the database on instance 1 is totally different than the one on instance 2.  When you use Datamaker, you have a connection profile for each database, correct?  With this in mind, it is my understanding that you would need Scramble installed on both.