Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Cloning DB to different UNIX platform

Re: Cloning DB to different UNIX platform

From: sg <s4v4g3_at_europe.com>
Date: Mon, 19 Aug 2002 01:00:25 +0200
Message-ID: <3D602709.7050904@europe.com>


as sybrand said, you dont know what you are talking about :)

nls_lang is one thing, database characters set is another

if you get characters set problems it doesnt mean ur nmls_lang is always bad, rather the database character set dont match or is not superset of other

statistics is always generated, that's why we disable them to speed up import process

objects do no get imported to tablespace system as long as they was not created in system in origin

as you can seee everyone who replied is telling you that your ideas aree wrong so I would point (again) to read server utilities manual and carry out some tests

we once had to migrate a 100GB SAP R/3 from aix to solaris and the only way was imp/exp

Rick Denoire wrote:

> Sybrand Bakker <postbus_at_sybrandb.demon.nl> wrote:
>
>
>

>>>Even doing the export with the appropriate option, statistics must be
>>>generated in an additional step after the import. Besides,
>>>export/import gives you at times some headaches with character sets,
>>>index names, constraints, triggers...
>>>
>>Statistics will be generated by imp when directed too
>>

>
> Not according to my experience. I was carefull at this point, but I
> might have done something wrong.
>
>
>>Problems with character sets can be prevented by setting an appropiate
>>NLS_LANG.
>>

>
> I onced tried all possible combinations of the NLS_? parameters,
> always got a warning (I tried at import only).
>
> Thanks
>
> Rick
>
Received on Sun Aug 18 2002 - 18:00:25 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US