RE: rman online backup with exclude undo tablespace

From: Mark W. Farnham <mwf_at_rsiz.com>
Date: Fri, 18 Jan 2008 11:50:27 -0500
Message-ID: <014601c859f2$3b8886e0$1100a8c0@rsiz.com>

Someone else has already mentioned ways to attempt to salvage data, but it seems you're not broken in production from what you've written. If you have an old enough copy of this file around and the intervening archived redo logs, you should be able to recover.  

Likewise, you can just take a new backup properly including the undo. If you're in a hurry to get the test system running, recovery of the file seems reasonable, but it is well worth your time to create and practice a working "disaster recovery scenario" without having to apply spit and bailing wire at the end of the process. Circa 1988 great care was required to get Oracle recovery "just right." For at least a decade is has been quite routine and the only complexity is figuring out which modes of recovery best fit your hardware, environment(s), and your requirements.  

I'm curious where you got the idea that you did not need the undo tablespace. If is was merely lack of understanding of recovery (restarts except for carefully constrained normal shutdowns routinely require some rollback after the redo is applied to account for transactions in flight that had database blocks flushed prior to the commit), that is one thing, but if someone is giving this as advice in developing a recovery or cloning protocol we need to debunk the notion. It would not be the first time anti-Oracle forces have spread pernicious rumors about how Oracle operates, and it would also not be the first time someone just plain wrong was giving out advice.  

Regards,  

mwf  

---

From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Prasad
Sent: Friday, January 18, 2008 9:39 AM
To: oracle-l
Subject: rman online backup with exclude undo tablespace  

All,

we had a online backup taken on a 10gr2 production db on solaris with configure exclude undo tablespace . we are trying to do a disaster recovery scenario to build a test system.

However we are getting into this scenario.

ORA-00704: bootstrap process failure
ORA-00604: error occurred at recursive SQL level 2
ORA-00376: file 2 cannot be read at this time
ORA-01110: data file 2: '/datastore/sfods03/undotbs01.dbf'

Thu Jan 17 20:14:19 2008
Error 704 happened during db open, shutting down database USER: terminating instance due to error 704 Instance terminated by USER, pid = 21434

Is it recoverable?

TIA
-Prasad

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

Received on Fri Jan 18 2008 - 10:50:27 CST