Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Mailing Lists -> Oracle-L -> RE: A special way of migration

RE: A special way of migration

From: Lex de Haan <>
Date: Wed, 28 Sep 2005 19:05:59 +0200
Message-Id: <>

"The whole export will be read into undo segments" is wrong; also, the export will *not* result in more undo generation or retention.

the export job "just" risks ORA-1555 (snapshot too old) in case concurrently running transactions have overwritten undo info needed to reconstruct read consistent block images. only the blocks dirtied since the export job started will be cloned in the buffer cache, and *only then* certain undo information is needed to reconstruct older block images.

so what you need to do (as Mark suggests below) is to size your undo tablespace and to set the undo retention appropriately. in 10g, you can set the retention to be guaranteed; until 9i, it is based on best effort.  



Steve Adams Seminar

-----Original Message-----
From: [] On Behalf Of Mark Bole
Sent: Wednesday, September 28, 2005 18:12 Cc:
Subject: Re: A special way of migration

I started a reply on this same point, then cancelled it, but since you (Michael McMullen) brought it up, I will add that the whole quoted paragraph below about "consistent=y" is a jumble of misinformation.

I suggest that over-use of sarcasm intermingled with technical information is resulting in less signal and more noise.

Exp/imp is the only way to perform a logical backup, and as such has its legimate uses, and should not be dismissed out-of-hand. Data pump in version 10g addresses some of the limitations of exp/imp that previously could only be addressed at the OS level (such as using pipefiles in Unix).

Strong arguments have been made that the growth of undo space should never be used as an excuse to break up a single business transaction. Make it as large as it needs to be to handle your longest-running transaction. Whether that is a consistent export or something else is not relevant.

-Mark Bole

Michael McMullen wrote:

> quote
> "The whole export will be read into undo segments."
> Were you joking on this?

Mladen Gogala wrote:
> On 09/28/2005 03:24:39 AM, Andre van Winssen wrote:


Received on Wed Sep 28 2005 - 12:08:10 CDT

Original text of this message