Hardware migration Non-Unicode source system to Unicode Target system

[Former Member]

Hello,

I am in the middle of a hardware migration of my Production CRM 5.0 (WAS 7.0) system. The current production system is non-Unicode. The migration path I have chosen is to install a new CRM 5.0 system on the new hardware and perform a database copy by means of detach/attach.

I have performed this type of migration several times and know the process quite well.

My issue is around Unicode/Non-Unicode. The new CRM 5.0 system has been installed using CRM 5.0 SR3 software and as a result is a Unicode system. When I perform the detach/attach of the live database will this automatically turn the new system into non-Unicode? I will also be changing the Kernel to non-Unicode.

I am aware of a couple of DLLu2019s in the system32 folder such as LIBRFC32.DLL

My platform is x64 Windows 2003Enterprise with SQL2005 DB.

Has anyone performed such a migration before and if so know of any issues?

[JPReyes]

That is described on the guide, check the Homogeneous system copy with unicode conversion.

Regards

Juan

[Former Member]

Hi Juan,

Thanks for your response.

I do not wish to perform a Unicode conversion. I want to overwrite my New (Unicode) installation, with the (non nunicode) Database and kernel of my current live system. Other than replacing the DB and Kernel are there any other steps to turn my New Unicode system back to non-unicode?

Regards,

Andy.

[JPReyes]

why don't you just install the same realease that you used in your original system and patch it up?...

regards

Juan

[Former Member]

Hi,

I wish I could, it would then be very simple. The system was originally installed as SAP CRM 3.1. It has been upgraded to 4.0 and then 5.0. The only install software available is CRM 5.0 SR3 (Unicode only) SR2 and earlier software had the option of unicode/non-unicode.

Regards,

Andy.

Edited by: Andrew Worthington on Nov 26, 2009 3:19 PM

Edited by: Andrew Worthington on Nov 26, 2009 3:19 PM

Edited by: Andrew Worthington on Nov 26, 2009 3:20 PM

[Former Member]

I think you need to install the system using sapinst -> system copy -> target system. There you still have the option to install non-unicode. And you should be able to copy the database then.

You cannot have a unicode database/kernel without doing the unicode conversion.

Cheers Michael

[JPReyes]

All the new releases come as Unicode you might have to bite a bullet and do the conversion, theres not really disadvantages beside having to go through the conversion. Beside that is a production system so read the hardware migration guide as your system will not be supported if its not done according to the guidelines.

Regards

Juan