RE: Unicode data migration from SQL Server to Oracle

From: Michael Rosenblum <mrosenblum_at_dulcian.com>
Date: Fri, 7 Nov 2008 16:17:19 -0500
Message-ID: <001001c9411e$37afc420$9300a8c0@dulcian.local>


Before shifting this topic to gateways/HS discussion...  

Back to the original question - I WAS using heterogeneous services and got a problem moving Unicode data from SQL Server to Oracle.  

Any help?  

Thanks,
Michael Rosenblum

-----Original Message-----

From: Jan Benjamins [mailto:jan.benjamins_at_gmail.com] Sent: Friday, November 07, 2008 4:04 PM
To: Richard.Goulet_at_parexel.com
Cc: mrosenblum_at_dulcian.com; Jared Still; oracle-l_at_freelists.org Subject: Re: Unicode data migration from SQL Server to Oracle

Hi,

You could look into oracle heterogeneous services, just google for these terms, you find enough, no need for transparent gateway (license). I think you need odbc for this, so for windows no problem, for unix you could look at the datadirect drivers (from Merrant ?). Used it years ago, pretty straighforward.

regards,

Jan

On Fri, Nov 7, 2008 at 7:11 PM, Goulet, Richard <Richard.Goulet_at_parexel.com> wrote:

Michael,  

    If that's what your developers want then your only option, that I know of, is Oracle's transparent gateway to Sql*Server. You install it on the Sql*server database machine & then create database links inside Oracle to it. Not very hard to set up(should take you about an hour max). Very easy to use and you can control whether or not they can update Sql*Server data from Oracle.  

Dick Goulet
Senior Oracle DBA
PAREXEL International
978.313.3426
 information transmitted in this communication is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please destroy any copies, contact the sender and delete the material from any computer.  


From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Michael Rosenblum
Sent: Friday, November 07, 2008 1:06 PM
To: 'Jared Still'

Cc: oracle-l_at_freelists.org

Subject: RE: Unicode data migration from SQL Server to Oracle

It is not a problem to push data from SQL Server to Oracle even using Microsoft's "Import and Export Data" utility (as long as Oracle's own ODBC drivers are used)  

But our developers want to have views on Oracle side that would point to SQL Server tables - and write migration scripts againt these views.  

Regards,
Michael Rosenblum
Dulcian Inc

-----Original Message-----

From: Jared Still [mailto:jkstill_at_gmail.com] Sent: Friday, November 07, 2008 12:12 PM To: mrosenblum_at_dulcian.com
Cc: oracle-l_at_freelists.org
Subject: Re: Unicode data migration from SQL Server to Oracle

Have you tried using the Migration Workbench in SQL Developer? It doesn't require using ODBC.
The Oracle connection is via Java connector, and the SQL Server connection is via open source SQL Server connector (jTDS)

If you do try it, use version 1.2. The new 1.5 version of SQL Developer doesn't work for the SQL Server -> Oracle migration.

I just used it this week, and have a question in on OTN for the 1.5 version.

Jared Still
Certifiable Oracle DBA and Part Time Perl Evangelist

On Fri, Nov 7, 2008 at 7:39 AM, Michael Rosenblum <mrosenblum_at_dulcian.com> wrote:

Good morning, all!  

Currently we are building a system in Ethiopia - and as a part of the project need to do a data migration. So.  

  1. Source - SQL Server 2000 Database with a number of tables with NVARCHAR datatype
  2. Stored data is in Ge'ez language - it does mean that there is no standard locale.
  3. SQL Server internally is using UCS-2 to store Unicode data (compatible with UTF16)
  4. Target - Oracle 10g (10.2.0.4) Enterprise edition on Windows 2003 Server 32-bit
  5. NLS_CHARACTERSET = AL32UTF8
  6. NLS_NCHAR_CHARACTERSET=AL16UTF16
  7. Migration mechanism - Heterogeneous Services via ODBC

Problem - Unicode data comes to Oracle as '?'.  

And it is not a visual affectation:  

SQL> select ascii("Column1") a, "Column1" from test_at_db_link A Column1
--------- --------

63 ?
SQL>   Attempted actions:

1. OLEDB - the same problem
2. Third party ODBC drivers - don't even recognize NVARCHAR columns
3. Played with HS_LANGUAGE and HS_NLS_LANG - no impact.
 

Any ideas/suggestions?  

Thanks,
Michael Rosenblum
Dulcian Inc.    

--

http://www.freelists.org/webpage/oracle-l Received on Fri Nov 07 2008 - 15:17:19 CST

Original text of this message