Doubt about unicode conversion

[Former Member]

Dear Friends,

I have a doubt about the unicode conversion. I have read (one server strategy) that I have to do export (R3load), delete the ECC 6.0 non-unicode, install an unicode database and then import with R3load. My doubt is about the deletion of ECC 6.0 non unicode. Why? can I convert to unicode only with an export (R3load) and subsequent import (R3load) without any deletion?. Obviously I must to change the SAP kernel with a unicode one. Is it correct?

Cheers

[Former Member]

you can not convert a SAP System to unicode by just exchanging the executables from non unicode to unicode ones.

a non unicode SAP Oracle database is typically running with a database codepage WE8DEC or US7ASCII (well this one is out of support).

so every string stored in the database is stored using this codepages or SAP-Internal codepages.

When converting to unicode you have to convert also the contents of the database to unicode. As unicode implementation starts at a time where Oracle did not support mixed codepage databases (one tablespace codepage WE8DEC the other one UTF8) inplace conversions are not possible. To keep things simpler, we still do not support mixed codepage databases.

Therefore you have to export the contents of your database and import it to a newly created database with a different codepage if you want to migrate to unicode.

regards

Peter

[Former Member]

Hi Peter,

many thanks for your help. I have another question for you. I would like to know if the same considerations about mixed codepages are valid also for others DBMS such as DB2 V8.1.

Thank you in advance for your precious help.

Bye,

Bob

[Former Member]

yes