Re: Sanity Check

  • From: Dan Norris <dannorris@xxxxxxxxxxxxx>
  • To: Mark.Bobak@xxxxxxxxxxxx
  • Date: Fri, 10 Oct 2008 16:19:23 -0500

Chris,

I can attest that I did this successfully. It was a long time ago, on 9i, but it worked fine. It's essentially the same as manually archiving the log. Just to be specific, Mark said to "copy on-line redo" from prod, but you only need to copy the *current* online redo log (that was current at time of shutdown) and apply it. Basically, just rename it to the next archived redo log in your log sequence (since that will likely avoid confusion).

I also agree that a standby database and a switchover would be better, but if you're using standard edition, that won't be an option.

Dan

Bobak, Mark wrote:

Chris,

 

To be sure, after copying all the archivelogs from the old server to the new server and applying all of them, take copies of the online redo logs, and apply them in the recovery process as well.  This will guarantee no data loss.

 

So, set up copy of prod on new hardware.  Start copying archive logs from prod to copy on new hardware and applying.  At cut over time, shutdown prod, copy remaining archive logs, apply them.  Then copy on-line redo from prod over to new hardware, and apply them as if they were archived logs (Oracle will not know the difference).  That will guarantee every last transaction will be applied to the new database.  Then just open w/ resetlogs and be on your way.

 

Hope that helps,

 

-Mark

 

--
Mark J. Bobak

Senior Database Administrator, System & Product Technologies
ProQuest
789 E. Eisenhower, Parkway, P.O. Box 1346
Ann Arbor MI 48106-1346
+1.734.997.4059  or +1.800.521.0600 x 4059
mark.bobak@xxxxxxxxxxxx
www.proquest.com
www.csa.com

ProQuest...Start here.

-- //www.freelists.org/webpage/oracle-l

Other related posts: