Re: Full Ex/Im-port to New Instance, NT

On Mon, 12 Oct 1998 08:17:43 +0100, "Yassir Khogaly" <yassir_at_khogaly.freeserve.co.uk> wrote:

>Oracle Corporation Support
>Microsoft Servers and Utilities Technical Forum.
>
>Displayed below are the messages of the selected thread.
>
>From: Charles Cooper
>Subject: Export/Import problem with Oracle 7.3 and Win NT 4.0
>Posted on: 12-Feb-1998 07:35:47
>
>RDBMS Version: 7.3.3
>Error Number (if applicable):
>Operating System and Version: Windows NT 4.0
>Tool (i.e. SQL*Loader, Import, etc.): Export/Import
>Tool Version: 7.3
>
>I am moving a PeopleSoft Financial database from
>one machine to another using a full export of the
>database and reorganising tablespaces etc. during
>the process. I have created the database instance
>and all the new tablespaces on the destination
>machine in their new locations. However, despite
>using FULL=Y on the export, when I run the full
>import, the user SYSADM does not get created and
>thus all PeopleSoft objects fail to import (since
>this is the schema in which they reside). Am I
>doing something wrong? Is there a limitation of
>export that I'm hitting?
> ... [SNIP]....
This is not a general limitation of import, it is a bug no.691338, fixed in 8.0.5

This bug happens if user performing an import (SYS) happens to have the id which is identical to users it is trying to import (SYSADM, SYSREP,...). The simple workaround for releases prior to 8.0.5 is to precreate users with this prefix, or even better (especialy if you happen to have many users starting with SYS or SYSTEM), precreate a single users with unique (in regards to prefixes of users being imported) username, like 'FULL_IMP_USER', give him appropriate privileges to perform full import, and run full imp under that user.

In no way should this bug force anyone to accept the practice to generaly precreate users, synonyms, grants etc prior to performing full import! This would simply be rediculous. Imp/exp have many more serious bugs in various releases when particular conditions are met, yet it would be unreasonable to advice people not to use imp/exp because of this bugs! At last, that's what bugs are for .... ;-)

Regards,

--
Jurij Modic <jmodic_at_src.si>
Certified Oracle7 DBA (OCP)
================================================
The above opinions are mine and do not represent
any official standpoints of my employer